About Structure Symmetry

M

miralay

New Member
Hello, I wish you good work. The mirror (miror) of a project we have done will be built on the same parcel. In order to get the drawings, I modeled the same building with the miror command, layer by layer. This is how I analyzed it in mirror form. Everything is exactly the same except that the two buildings are mirrors of each other. As a result of the analysis, I observed differences in cross-section effects. In summary; 1-) In vertical load combinations (G, G+Q etc.), the cross-section effects (N, V, M) are largely the same (approximately 99.9%). I think that this small difference may be due to the differences to be made when installing the finite elements. 2-) There are serious differences in combinations with earthquakes. There are elements that come out 20-30% or even more in rank. Isn't it expected that the static analysis results of a structure that is a logical mirror will be the same? Am I thinking wrong? When comparing 2 models, especially in combinations with earthquakes, shouldn't the G+Q+Ex combination in the 1st model be the equivalent G+Q-Ex in the 2nd model? I can't upload here because my models are too big. However, if I try it on a small model, the result is the same. Is it wrong to produce a model with the miror command?
 
Hi, you can generate symmetry using the K. Mirror command. If the structure created with the symmetry command or by entering information is symmetrical in terms of both geometry and loads, all results are symmetrical. The seismic forces applied to the structure are in the X and Y directions. If the structural symmetry is taken in the direction of the X or Y axes, it is expected that the extreme forces resulting from the earthquake loading will be the same. If you are getting the structure symmetry in the X or Y axes and the results are not symmetrical, if you send the project files to [email protected], we will examine the situation. Good work
 
Mr. Levent, good work. I can't upload my own project because of the file size. However, I made a very simple small model for testing purposes. In the appendix, Model1 and Model2 are exactly the same building, multiplied from the Y axis with the symmetry command. Examine the cross-sectional effects of any column on the ground floor in combinations with earthquakes. Link of Files
You do not have permission to view link Log in or register now.
I wish you good work.
 
Good work, in the picture below, the cross-sectional effects of the S08 column in both models were examined in combinations of G+Q+Ex and G+Q-EX, which in my opinion should give the same results. Minor directional moments and torsional forces are the same. However, there is a 2 times difference in axial load.
You do not have permission to view link Log in or register now.
You do not have permission to view link Log in or register now.
 
When these small projects you sent are examined, it is seen that the extreme forces of the project produced by taking the symmetry are exactly the same as the other project. 1) All results are 100% the same. There is not even a very small difference. 2) Earthquake results are also 100% the same. There is not even a very small difference. 3) If you do a response spectrum solution (as here), your suggestion is not correct since all results will be positive (ABS). This is valid for static horizontal loads.
 
I also added axial load charts. I explained it by making a small addition and subtraction below. When I open and examine the models, all the results are consistent, accurate and as they should be. Let's examine the combination of G + Q + Ex and G + Q - Ex. In the picture below there is G + Q : In the picture below there is Ex In order to find the most unfavorable result in response spectrum combinations, the value ( Ex ) calculated as an absolute value is added to G + Q once as positive and once as negative. Sum the combination of Model1 Ground floor S01 column and Model2 S01: Model1 G + Q value Lower End -38.272 tf Model1 Ex value Bottom end +20.481 tf G + Q + Ex = -38.272 + 20.481 = - 17.791 tf G + Q - Ex = - 38.272 - 20.481 = -58.791 tf Model2 G + Q value Bottom End -38.272 tf Model2 Ex value Bottom end +20.481 tf G + Q + Ex = -38.272 + 20.481 = - 17.791 tf G + Q - Ex = -38.272 - 20.481 = -58,791 tf
 
Mr. İsmail, how does this situation explain the axial load difference in both models? Is it wrong to compare two symmetrical models in combinations with earthquakes with different signs?
 
There is no difference in axial load between the two models. In this system, the columns that are symmetrical with respect to the Y axis are S01 columns. Compare Model1 Ground floor S01 column with Model2 S01 columns: Model1 G + Q value Lower End -38.272 tf Model1 Ex value Bottom end +20.481 tf G + Q + Ex = -38.272 + 20.481 = - 17.791 tf G + Q - Ex = - 38.272 - 20.481 = -58.791 tf Model2 G + Q value Bottom End -38.272 tf Model2 Ex value Bottom end +20.481 tf G + Q + Ex = -38.272 + 20.481 = - 17.791 tf G + Q - Ex = -38.272 - 20.481 = -58,791 tf
 
Mr. İsmail, if the combination we are looking at is the combination G+Q+EX, the axial loads of the S01 Columns are the same. But this time the Moment and shear forces are different. According to the same logic, wouldn't it be expected that the torque and shear forces values will be the same?
You do not have permission to view link Log in or register now.
You do not have permission to view link Log in or register now.
 
It is not expected to appear in the same combination for this structure. What should come out is understood by checking manually. Let's manually check the moment combination you mentioned for the same column: G + Q Moment M33 values Ex Moment M3 values Model1 G + Q value Lower end +0.506 tfm Model1 Ex value Lower end +10.682 tfm G + Q + Ex = + 0.506 + 10,682 = +11.188 tfm G + Q - Ex = + 0.506 - 10,682 = -10.176 tfm Model2 G + Q value Bottom End - 0.506 tfm Model2 Ex value Bottom end +10.682 tfm G + Q + Ex = - 0.506 + 10,682 = + 10,176 tfm G + Q - Ex = - 0.506 - 10,682 = - 11,188 tfm As can be seen, S1 column M33 moments are calculated and compared manually G+Q +Ex == G+Q-Ex is happening. In the program, he finds these values in these signs. What is done in combinations in the program consists of adding the values, taking into account their signs. G + Q + Ex Model 1 System on the left +11.188 tfm calculated by hand M33 G + Q + Ex +11.188 tfm calculated by the program Model 2 System on the right +10.176 tfm calculated by hand Program calculated M33 G + Q + Ex +10.176 tfm G+Q-Ex Model1 Left system Calculated by hand -10.176 tfm Program calculated M33 G + Q - Ex -10.176 tfm Model2 System on the right Manual calculated -11,188 tfm Program calculated M33 G + Q - Ex -11,188 tfm
 
Mr. İsmail, in this case, when the "Apply horizontal floor forces calculated by combining modal contributions" option is selected in the analysis settings of the same models, the G+Q+EX combination of the 1st model and the G+Q-EX combination are exactly the same. So I think there is an error in the mirror command.
 
Dear Miralay, the ideCAD Mirror command works without any errors. G + Q results are exactly the same on both systems. Ex and Ey are exactly the same on both systems. The signs of these downloads are as they should be. Combination is an addition and subtraction operation. I tried to explain these processes clearly in the messages above. When the "Apply horizontal story forces calculated by combining modal contributions" option is selected, Ex and Ey loading end forces are marked. Ex is loaded once from left to right and once from right to left. My advice to you is to enter the Model 2 system you created by mirroring from scratch in the same way in terms of both geometry and loads. Name the system you entered independently as Model 3. Then compare Model 2 with Model 3. Model 2 and Model 3 will give the same result for all elements in all loadings and combinations. Solve Model 2 and Model 3 with both response and "Apply horizontal story forces calculated by combining modal contributions" option. Thank you for your interest
 
The case with a true axis of symmetry. The part on the right was duplicated with the mirror command. Here all the results of the structure on the left are the same as in Model 1 All the results of the structure on the right are the same as in Model 2 Here all the results of the column S01 of the structure on the left and the column S01 of the structure on the right are the same. Best regards
 
You must log in or register to reply here.
Back
Top