Symmetric Structure asymmetric reaction?

[TCen]

Hi everyone,

I have this simply symmetric structure (3 meter * 6 meter * 3 meter height) modelled using etabs.

I defined the diaphram for the slab and wind load Wx that applied on the diaphram, which is 8.94kN in X direction.

After running the model, I checked the reations under Wx and found that the reactions of 2 walls are different. I am very confused as these two wall are of same metrial & size.

Is there anything that I missed before running the model?

Thank you.

 

[JoshPlumSE]

Yes, there must be something you missed. However, it is impossible to tell from the information provided.

Maybe some adjustment factor on the walls to account for cracking or such. Maybe something about the diaphragm that ties the walls together. Perhaps a meshing issue with the walls or diaphragm. Or, maybe there is another load that we can't see.

 

[sbisteel]

Automatically generated accidental eccentricity in the diaphragm?

 

[WARose]

Just a guess.....but have you checked which direction the gravity self-weight is applied?

 

[rb1957]

these look like reactions to the overturning moment, assuming a lateral load applied to the diaphragm.

another day in paradise, or is paradise one day closer ?

 

[JLNJ]

Despite what your plan shows, it looks like your load is simply not centered in the Y direction.

 

[BAretired]

The applied load is 8.94kN at a height of 3m. The vertical reactions are 6.25 and 2.69 which add up to 8.94 as they should. The location of the XZ plane appears to be off center in the diagram which could account for the discrepancy if the applied load is located at Y = 0.

Why does your program not show horizontal as well as vertical reactions? Perhaps they are out by the same factor.

BA

 

[rb1957]

If the load is centered (looks like that on the left pic, in the right pic it looks like the axes are shown on the ground plane), then for some reason (not shown) one support is stiffer than the other (and so reacting more moment). The overall side view looks to be square, so the over-turning moment reactions are the same as the applied load (it least that's how it looks to me).

Please restate the problem clearly. Show the applied load, show the constraints, show the free body reactions, figure out if there is some unseen stiffness having an effect.

another day in paradise, or is paradise one day closer ?

 

[Settingsun]

I generally agree with the previous two posts (but am not familiar with your particular software). Are the reactions in equilibrium with the applied loads? You don't show any horizontal reactions as BAretired said.

Did you define everything as perfectly-rigid diaphragms (ie both the slabs and walls)? Are the supports also perfectly rigid? It could just be that everything is too rigid and it's essentially a random result (subject to being in equilibrium with the applied loads). Even though it's symmetrical and logically the results would be expected to be symmetrical, it's an indeterminate structure with many different possible solutions and relies on flexibility to find a solution.

Try making the supports stiff but not perfectly rigid. Say 100 kN/mm or 1000 kN/mm. That might force the high reaction to balance with the low reaction.

 

[rb1957]

if the reacting shear load does follow the pattern of the couple reactions (as it should, else why is the couple distributing itself that way ?) then either
1) the load is not applied at the mid span, but is off-set towards the higher reactions, or
2) there's another couple at the reactions, balancing the moment (from having the load off-set from the reactions). Why would a structure do this ? if one side is much stiffer (about 3 times) than the other.

another day in paradise, or is paradise one day closer ?