The coefficient of torsional irregularity is very close to 1

[emrahgokdemir]

Hello there. I designed a project. I present the relevant model as downloadable in the attached link. After the analysis, both the relative storey drifts and the torsional irregularity coefficients remained below the limit. But the torsional irregularity coefficients were very close to 1. (Values like 1.002-1.005. maxu 1.042) I suspected that the values were so close to 1. Because the coefficient is 1, it means that all the columns in the relevant floor have almost equal relative story offsets. (Correct me if I'm wrong, please) I checked my modeling again. But I couldn't find any serious modeling error. Suspicious of torsion coefficients, I changed all the shears in the building to 30/50 columns. However, the torsional irregularity coefficients remained almost unchanged. Then, I started to lift columns from the eastern part in order to weaken the east relative to the west, relative to the center of mass of the building, so that the building would exhibit torsion behavior. Even though I removed all the columns on the east side (even when I tried to carry an area of approximately 100 m2 with 1 single column), the torsional irregularity coefficients remained around 1.007. So I think there is a mistake in my project, but I can't find where I made a mistake. I would be very grateful if you could review my model and help. MODEL DOWNLOAD LINK:

You do not have permission to view link Log in or register now.

Best regards, Emrah GOKDEMIR

 

[nurgulkaya]

Re: THE COEFFICIENT OF TORSIONAL IRRORITY IS VERY CLOSE TO 1 Hello; Torsional irregularity control is calculated by Equivalent earthquake load method as per TBDY 2018 Table 3.6. Relative story drift and internal forces are made according to MODAL analysis, if the forces that are essential to the reinforced concrete design are. Therefore, the results will not be exactly the same. The reason for the situation in your project is due to the logic in the selection of semi-rigid. In semi-rigid diaphragm, depending on your mesh spacing, n nodal points are obtained in slabs. And the building mass is distributed to the nodes in these floors. Even if there is an irregularity problem, there is no problem since the calculation is made with the dynamic analysis method as stated in the related torsional irregularity page in the report. You can turn it to a fully rigid diaphragm and check for irregularity. There does not appear to be a modeling problem in your project.

 

[emrahgokdemir]

Re: THE COEFFICIENT OF THE TORSIONAL IRRIGATION COEFFICIENT IS VERY CLOSE TO 1 Thank you for your reply. However, the fact that the torsional irregularity coefficients are so close to 1 (almost 1) is not very plausible in my opinion. Because, in order for the coefficient to be 1, the column values with Max reduced relative story drift must be the same as the column value with the minimum relative story drift, so that the ratio is 1. Therefore, the reduced relative storey drift values of all columns in the relevant storey must be equal to each other. In my model, although the values are not 1, I am getting values very close to 1. (like 1.003-1.005-1.007). As a result of my long research to determine the source of the problem, I came to the following conclusion; -I removed the K25 and K16 stud beams on the model. Therefore, I removed the 85*155 slab tear, which corresponds to some of the P01 and P05 curtains, and combined the D23-D24 slabs and D04-D13 slabs among themselves and passed as a single slab. (In summary, I have removed 2 simple stud beams and floor tears) -When I analyzed the model without any other changes, I observed that the torsional irregularity coefficients and relative storey drifts increased significantly. (The torsional irregularity coefficients reached from max 1.078 to max. 1.588.) (The relative storey drifts reached from max. 0.00666 to max. 0.01119.) As a result; It doesn't make sense for the values to change so dramatically with this small change I've made. Because in the first model, no matter how many columns I decrease, the torsional irregularity values do not change much, they remain very, very close to 1. However, after my second modeling, the changes I made in the columns are also reflected in the torsional irregularity coefficients in parallel with the change. -I think that such a problem occurred due to 85*155 tears, which corresponds to some of the P01 and P05 Curtains in the first model. I request that the problem be examined in detail and that if the problem is software-related, it is solved, and if it is not software-related, I request that the problem and its solution be revealed. Note: I am adding screenshots for better understanding of the subject. I am attaching the download link of both Models mentioned. You can download the models and get the results I wrote. DOWNLOAD LINK:

You do not have permission to view link Log in or register now.

 

[emrahgokdemir]

Re: TORSIONAL Irregularity COEFFICIENT IS VERY CLOSE TO 1

 

[nurgulkaya]

Re: THE COEFFICIENT OF TORSIONAL IRRORITY IS VERY CLOSE TO 1 Hello; The removal of the stud beam and tear is a situation that changes the analysis model. Therefore, we do not expect the results to be the same anyway. The situation in your project has been sent to the R&D team for review. A1 irregularity control is only effective in deciding whether to choose the equivalent earthquake load method or modal analysis as the analysis method. Since ideCAD Static already calculates with modal analysis, it is not important. However, it is necessary to check for relative offset. Until the R&D team returns, we recommend that you open the program with ctrl +shift and re-enter it so that there is no problem in the sets or intermediate steps you created your project. Our recommendation is to connect the two curtain groups with beams in the middle area and arrange the spaces. It is suitable for the curtains to work together in the middle region and the formation of a rigid core region will seriously change your results and will be a more suitable result in terms of engineering.

 

[emrahgokdemir]

Re: TORSIONAL IRRORITY COEFFICIENT IS VERY CLOSE TO 1 Well, while there are stud beams and tears in the first model, can you try to remove all 30/50 columns to the east of the building according to the center of mass? Even if you remove all 6 columns in the eastern region, even if you try to carry that area entirely with beams, you will observe that the torsion coefficients in the structure are still very close to 1 and do not change much. How normal is this? When we weaken the east of the structure more than the west, wouldn't it be expected that the torsional irregularity coefficients will change drastically? I understand what you wrote, but unfortunately my logic does not accept this situation.

 

[emrahgokdemir]

Re: TORSIONAL Irregularity COEFFICIENT IS VERY CLOSE TO 1 Although this is not our topic, "A1 irregularity control is only effective in deciding whether to choose equivalent earthquake load method or modal analysis as analysis method. ideCAD Static is already modal It doesn't matter because it calculates with analysis." you wrote. The nbi coefficient we found in A1 irregularity remains in the range of 1.2-2.0, it doesn't matter if the dynamic method is used, you are right. But it's not limited to that, you know. In accordance with the regulation, the eccentricity ratio should be increased by taking the max nbi coefficient and multiplying the coefficient obtained from the (nbi,max /1.2)^2 formula with the eccentricity ratio of 5 percent. Therefore, since nbi,max value is 1.078 in the first model, my eccentricity value will be 5 percent. However, since my nbi,max value is around 1.580 in my second model; (1.580/1.2)^2=1.73 coefficient will be increased in the eccentric, so it will be 5% * 1.73 = 8.65. Therefore, the change in the torsion coefficients will cause the eccentricity to increase by around 70 percent. This will cause the system to be exposed to higher torsional moments due to high eccentricity at -5%X, +5%X, -5%Y, +5%Y loadings. In addition, since the coefficient is greater than 1.2, the B coefficient will increase from 0.8 to 0.9. Therefore, if there is a software error in the calculation of the torsional irregularity coefficients, I think that this will significantly affect the structural design. Good work Best regards, Sent from my SM-G770F using Tapatalk

 

[nurgulkaya]

Re: THE COEFFICIENT OF TORSIONAL IRRORITY IS VERY CLOSE TO 1 Hello; The calculation and tests of this coefficient were applied with ETABS and comparative models. These tests are the BA1 example etc. in the IMO handbook. It was made for structures in geometry. We think that it is project-based or caused by a situation in your structure. It doesn't have to be your fault or the program's fault. There may be a situation related to account acceptances. For this, the necessary notes have been created in the system. Good work.

 

[emrahgokdemir]

Re: THE COEFFICIENT OF THE TORSIONAL IRRIGATION COEFFICIENT IS VERY CLOSE TO 1 Thank you for your attention. I wish you good work. Sent from my SM-G770F using Tapatalk

 

[goose]

Re: THE COEFFICIENT OF TORSIONAL IRRIGATION COEFFICIENT IS VERY CLOSE TO 1

But you know it's not limited to this. The eccentricity ratio should be increased by multiplying the eccentricity ratio of 5 percent by the coefficient obtained from the formula. If there is no point, the torsional irregularity does not have any direct effect on the mode coupling calculation in terms of regulation.

 

[emrahgokdemir]

But you know it's not limited to this. In accordance with the regulation, the eccentricity ratio should be increased by taking the max nbi coefficient and multiplying the coefficient obtained from the (nbi,max /1.2)^2 formula with the eccentricity ratio of 5 percent.

Isn't this the 4.7.4 article you wrote? If it is, it concerns the equivalent earthquake calculation, it does not need to be increased when the calculation is made with the mode combination method. If there is no point that I have missed at the moment, torsional irregularity does not have any direct effect on the mode combination calculation in terms of regulation. It is only useful to give an idea about the structure.

You are right. However, even if the Mod combination method is used, in accordance with Article 4.8.4.1, if the nbi coefficient is greater than 1.2, the katsayısıE coefficient used in the calculation of the B coefficient will increase from 0.8 to 0.9. The found coefficient B will also be taken into account as a multiplier in increasing all the reduced internal forces and displacements. Therefore, even this situation alone will cause an increase in internal forces and displacements by 12.5%. If you download and examine the models I have added, the torsion coefficients almost do not change and come very close to 1, no matter what you change in the system with the first slab tear (column reduction, column enlargement-reduction, etc.). But when you remove the small tear in the floor, the torsion coefficients increase to the level of 1.500, and the system changes the torsion coefficients in parallel to all changes made after that moment (such as column lifting, column enlargement and reduction). Yes, it is normal for the results to change, as removing the small tear in the flooring will change the analysis system. But change the place you want in the system before the tear in the floor is removed (so change the analysis model as you wish); The system does not change the torsion coefficients in parallel with the change you made and the coefficients are still very close to 1, almost the same. I think that by removing this tear, the results will change in parallel with the changes made; but when the tear is not removed, it is very unreasonable to remain insensitive to the changes in the calculation of the torsion coefficients. I haven't been able to pinpoint the source of the problem, but it's obviously not plausible. Sent from my SM-G770F using Tapatalk

 

[goose]

Yes, you are right too, when I focused on the 1.2 formula you wrote, the item 4.8.4.1 did not come to my mind.

 

[Ismail Hakki Besler]

If your projects have unexpected results like this, it may have something to do with the mesh ranges you selected. For example, if the finite element width is selected as 20 cm for both curtains and floors in this project, this situation may not occur. You chose 60 cm for curtains and 40 cm for upholstery. In general, it may be useful to take the risk of doing some long-term solutions and not choose mesh intervals smaller than 20 cm. ideCAD produces finite elements duly and automatically in any project. However, such large mesh ranges may not always yield appropriate results when selected. When you send us these situations that occur when the mesh ranges are selected large, we take the importance of not creating mesh, which is also necessary in cases of large mesh ranges. In addition, when the mesh intervals are 60 cm, we can understand why this situation occurs and we can find the solution. In this sense, sending a project helps us develop accurate solutions to problems in a much shorter time. Another point is that choosing large mesh intervals does not always lead to similar situations in every project. We will examine your project in the R&D department.

 

[emrahgokdemir]

If you have unexpected results like this in your projects, it may be related to the mesh ranges you selected. For example, if the finite element width is selected as 20 cm for both curtains and floors in this project, this situation may not occur. You may not expect 60 cm for curtains. you have chosen 40 cm for tiles. In general, it may be useful to take a little long-term solution and not choose mesh intervals smaller than 20 cm. ideCAD produces finite elements duly and automatically in any project. However, it may not always give appropriate results when such large mesh intervals are selected. When you convey these situations that occur when the mesh ranges are selected large, we take the importance of not being able to create mesh, which is also necessary in cases of large mesh intervals. We can also understand why this situation occurs when the mesh ranges are 60 cm and we can find the solution. In this sense, sending a project will help us develop accurate solutions to the problems in a much shorter time. happens. Another point is that choosing large mesh intervals does not always lead to similar situations in every project. We will examine your project in the R&D department.

Mr. İsmail, thank you for your interest. I switched to ideCAD from a different static calculation program. Once again, I have confirmed that I made the right decision in choosing ideCAD. It is really admirable that you examine the issues in detail without being biased and that you do not take an attitude of looking down on your users and take our requests and suggestions into account. We, as your users, will try to show our support in order to develop and strengthen IDECAD day by day. You are truly a valuable asset to our country. I personally regret to have met the ideCAD family so late. Good work, Emrah GÖKDEMİR Sent from my SM-G770F using Tapatalk

 

[MustafaAkın20]

Mr. Emrah, check the stair support conditions. All edges seem to be supported by columns, beams, slabs, curtains, and basic elements. Even that causes serious good/bad change. Nurgül hnm, like torsional irregularity, floor drift control is also done according to equivalent earthquake load, am I wrong?

 

[emrahgokdemir]

Hello, Mr. Mustafa, As far as I know, after entering the ideCAD stair support conditions command, click on any side of the ladder and select the elements (attached to columns, attached to beams, attached to curtains, etc.) that are marked in the "supporting conditions for solution with structure"; It searches all the edges at a distance of the entered support distance (for example, 25 cm), and finds the marked elements that are 25 cm away and links them to the relevant places. If there is no one of the elements marked in the said distance (25 cm) range, it does not perform a support operation for that element. Therefore, even if we have selected an element that is not around the staircase in the "supporting conditions for the solution with the structure" section, ideCAD will search for this element near the edges of the stairs, and will not support it because it cannot find such an element nearby. Therefore, I think that choosing all the options in the "supporting conditions for the solution with the structure" section will not cause a modeling and design error. Please correct me if I'm wrong about this.

 

[MustafaAkın20]

Hello, I have seen that it is necessary to mark the support conditions of the ladder manually, whichever element is supported, both in the information I received from the forum and in the experiments I made about the ladder. When I first used it, I set up a logic like you, but the important thing is which logic the program analyzes. Therefore, the user needs to define the support on the stairs. There are similar videos on the forum. take a look if you want.

 

[emrahgokdemir]

Hi, I saw that I needed to mark the support conditions of the ladder manually, both in the information I got from the forum and in my experiments. The first time I used it, I set up a logic like you, but the important thing is with which logic the program analyzes. Therefore, the user needs to define the support for the stairs. There are videos similar to this on the forum. Take a look if you want.

 

[emrahgokdemir]

If your projects have unexpected results like this, it may be related to the mesh spacing you have chosen. For example, if the finite element width is selected as 20 cm for both curtains and floors in this project, this may not occur. you have chosen 40 cm for tiles. In general, it may be useful to take a little long-term solution and not choose mesh intervals smaller than 20 cm. ideCAD produces finite elements duly and automatically in any project. However, it may not always give appropriate results when such large mesh intervals are selected. When you convey these situations that occur when the mesh ranges are selected large, we take the importance of not being able to create mesh, which is also necessary in cases of large mesh intervals. We can also understand why this situation occurs when the mesh ranges are 60 cm and we can find the solution. In this sense, sending a project will help us develop accurate solutions to problems in a much shorter time happens. Another point is that choosing large mesh intervals does not always lead to similar situations in every project. We will examine your project in the R&D department.

Hello Mr. Ismail, I solved the project by taking the finite element widths of 20 cm for floors and curtains as you said. However, there is no difference in the calculation of torsional irregularity and relative storey drifts. Then I solved it by taking the finite element widths to 10 cm. After about 40 minutes of analysis with 250000 nodes, there was no significant difference in the results I got either. I am sharing the results I got with a finite element width of 10 cm. The problem may not be related to mesh widths. For your information, Best Regards.

 

[proisa]

I am facing the same problem as Mr. Emrah in some projects. The most obvious parameter for us to realize the error is that the torsional irregularity is very close to 1.00 or 1 and the relative storey drifts of the building are lower than they should be. As someone who constantly draws projects, I can intuitively observe that the building is solved much more easily than it should be. In the project I added to the message, although only columns with minimum dimensions are used as a preliminary study, I do not get any errors other than double reinforcement error or joint zone shear safety error. In addition, the torsional irregularity coefficient is 1.00, except for the Basement Floor, when some random stud beams are deleted in sequence, the problem is corrected, but I could not detect that it is related to only one beam. I reanalyze each time I delete a stud beam. After I found the beam where the results were correct, I reloaded the project as it was and deleted that beam at first, but I could not reach the same result. After deleting a few more beams this time, the analysis results change. I couldn't find an error in my model. Maybe I have made a data entry error that I am not aware of. I wish you good work...

You do not have permission to view link Log in or register now.