Confusion regarding Horizontal Loads, Footings and Tie Beams:

[IngDod]

Greetings,
I have a confusion regarding the design of footings with pedestals when lateral loads (Vx and Vy)are present; specifically shear from seismic loads. I have a footing that receives axial and horizontal load from a pinned column, so in principle there is no moment transfer. However, the pedestal is 1m in height (Hp) and Im assuming that this introduces a moment to the footing equal to My=Vx*Hp; is this correct?. Also as per code requirements I must use a "Tie beam"... Basically a concrete member connected to the top of each pedestal; since there is no moment transfer I'm guessing I will design them only for axial load, some people have told me to design only for tension while others for both tension and compression.. I'm confused on this subject. Lastly, by having this tie beams should I still design the footings for the moment generated by Vx*Hp?; My first impression is that the tie beams prevent the pedestal from deflecting with the load and thus avoid generating moment while still transferring lateral load to the soil; but since the tie beams joins two or more pedestals it might move laterally as a whole and still generate moment on all footings due to the lateral load. Thanks.

 

[BAretired]

I have a confusion regarding the design of footings with pedestals when lateral loads (Vx and Vy)are present; specifically shear from seismic loads. I have a footing that receives axial and horizontal load from a pinned column, so in principle there is no moment transfer. However, the pedestal is 1m in height (Hp) and Im assuming that this introduces a moment to the footing equal to My=Vx*Hp; is this correct?.

It is not a "moment to the footing". There is an applied shear and moment at the top of footing. The moment at the middle of the footing is larger by Vx*t/2 and the moment at the bottom of the footing is larger by Vx*t where t is the thickness of the footing. Statics applies. Resisting the applied moment, there is soil pressure acting on the sides and bottom of footing.

Also as per code requirements I must use a "Tie beam"... Basically a concrete member connected to the top of each pedestal; since there is no moment transfer (Why is there no moment transfer between tie beam and pedestal? Are they not tied together with reinforcement? Tie beams resist pedestal rotation.) I'm guessing I will design them only for axial load (not true), some people have told me to design only for tension while others for both tension and compression (why not bending?). I'm confused on this subject. Lastly, by having this tie beams should I still design the footings for the moment generated by Vx*Hp?; (only if you pin the tie beams to the pedestal, but why would you do that?) My first impression is that the tie beams prevent the pedestal from deflecting (how do they do that, unless you are counting on friction between the soil and the tie beams?) with the load and thus avoid generating moment while still transferring lateral load to the soil; but since the tie beams joins two or more pedestals it might move laterally as a whole and still generate moment on all footings due to the lateral load. When two or more pedestals are connected by tie beams, they must move together. I suggest that you draw a sketch and apply statics. The problem should then become clear.

BA

 

[IngDod]

Thanks for your reply,I'll try to explain myself better:
I have a moment frame structure, the columns are HSS welded to baseplates (No stiffeners to prevent deformation of the HSS wall); my assumption in this case is that the connection behaves as pinned so there is no significant moment transfer from the structure to the foundations. Since there are lateral loads on the structure they are transferred as shear in the HSS column - footing pedestal interface; since there is no moment at the pedestal top my assumption is that there is no moment to transfer to the tie beams; thus they only take axial load. My question is regarding how the shear (Lateral Force) at the top of the pedestal interacts with the footing which is located at the bottom of the pedestal and the tie beams. My impression, from statics, is that the footing will receive a moment equal to the shear at the top of the pedestal times the pedestal height; now for this to happen the pedestal must be able to deflect horizontally (Like a cantilever beam with a load at its end), no deflection no moment.. My other question was: Given that the tie beams joins two or more pedestals they might provide some measure of horizontal restraint to each pedestal; whether its enough to assume no deflection occurs at each pedestal and thus no moment was my doubt; reading your answer I can see that all the pedestals and tie beams would move together.

Thanks to your answer i believe my error was in assuming that the tie beam would only take axial load because there was no moment coming from the hss column; however the rotation induced in the footing by the lateral force is resisted by the tie beam as a moment as well an axial load. I have some books saying that its possible to assume that the tie beams take the whole moment and so the footing can be designed for axial load only; I imagine that what really happens is that the moment in the footing is reduced (Not completely of course) by the transfer that occurs at the tie beam. Please correct me if I'm totally wrong; this has always been a doubt of mine. Would an analysis using a one story frame with fixed supports provide an approximate model of the behavior?; the columns would be the footing pedestals, the beams the tie beams and the fixed supports the pedestal - footing interfaces?. The moment at the footings would be reduced and they would take the whole lateral loads as reactions at the supports; I would have to check that sufficient friction between footing and soil develops. I imagine a model like this would be conservative since no soil friction is taken into account.

I was doing my design under the assumption that the footing takes the whole lateral load and moment, as if no tie beams were present. I was designing the Tie beams for 10% of the axial load in the column, as specified in various codes and books i have read.

 

[BAretired]

Thanks for your reply,I'll try to explain myself better:
I have a moment frame structure, the columns are HSS welded to baseplates (No stiffeners to prevent deformation of the HSS wall); my assumption in this case is that the connection behaves as pinned so there is no significant moment transfer from the structure to the foundations.

HSS columns with baseplate and four anchor bolts are not really hinged, but it is common practice to make that assumption, so I take no issue with it.

Since there are lateral loads on the structure they are transferred as shear in the HSS column - footing pedestal interface; since there is no moment at the pedestal top my assumption is that there is no moment to transfer to the tie beams; thus they only take axial load.

That is not a valid assumption, although you can do it that way if you wish. If you do, the pedestals are free to rotate. If you attach the beams with a moment connection, you reduce the pedestal rotation and the moment to be carried by the footing.

My question is regarding how the shear (Lateral Force) at the top of the pedestal interacts with the footing which is located at the bottom of the pedestal and the tie beams. My impression, from statics, is that the footing will receive a moment equal to the shear at the top of the pedestal times the pedestal height; now for this to happen the pedestal must be able to deflect horizontally (Like a cantilever beam with a load at its end), no deflection no moment..

The footing will receive a smaller moment if the tie beams resist part of it. I agree with your comment "no deflection, no moment" but you will get deflection unless the soil is infinitely rigid, so it goes without saying that the footing will receive some moment.

My other question was: Given that the tie beams joins two or more pedestals they might provide some measure of horizontal restraint to each pedestal; whether its enough to assume no deflection occurs at each pedestal and thus no moment was my doubt; reading your answer I can see that all the pedestals and tie beams would move together.

If all pedestals simultaneously receive a force in the same direction, as you might expect from wind or seismic load, they all tend to move in the same direction so the tie beam does nothing to prevent deflection of pedestals.

Alternatively, if some pedestals receive a southward force while others receive a northward force, then they tend to balance one another and tying them together definitely reduces lateral deflection.

Thanks to your answer i believe my error was in assuming that the tie beam would only take axial load because there was no moment coming from the hss column; however the rotation induced in the footing by the lateral force is resisted by the tie beam as a moment as well an axial load. I have some books saying that its possible to assume that the tie beams take the whole moment and so the footing can be designed for axial load only; I imagine that what really happens is that the moment in the footing is reduced (Not completely of course) by the transfer that occurs at the tie beam. Please correct me if I'm totally wrong; this has always been a doubt of mine. Would an analysis using a one story frame with fixed supports provide an approximate model of the behavior?; the columns would be the footing pedestals, the beams the tie beams and the fixed supports the pedestal - footing interfaces?. The moment at the footings would be reduced and they would take the whole lateral loads as reactions at the supports; I would have to check that sufficient friction between footing and soil develops. I imagine a model like this would be conservative since no soil friction is taken into account.

I think you are correct in the above.

I was doing my design under the assumption that the footing takes the whole lateral load and moment, as if no tie beams were present. I was designing the Tie beams for 10% of the axial load in the column, as specified in various codes and books i have read.

I am not aware of the 10% rule and can see no theoretical basis for it. Ultimately, lateral forces are taken by a combination of shear between footings and soil and by passive pressure between soil and pedestals or attached elements below grade such as elevator shafts or exterior walls. Without considering specifics, it is difficult to comment intelligently on precisely how the horizontal forces are resisted.

BA