RAM Concept Self Equilibrium Seismic loads

I'm having difficulty with a model that I'm integrating between a RAM SS model and RAM Concept for a PT slab design.  I successfully designed my Level 4 floor slab and got it working well. To design Level 3, I copied all my Design Strips, Prestressing, etc. and got it working relatively quickly.

After I did the same thing for Level 2, I started getting ridiculous answers... Even when the load balancing was the same, it was telling me some spans that worked on Level 3 and 4 with (1) #5 bar added now required over (300) #5 added in Level 2 (and similarly ridiculous problems in shear, etc). I finally isolated that the problem had to do with the seismic loading, and now know it has to do with the Lateral SE loading. I can run the Concept design with the seismic load cases set to "Normal" and everything is nearly identical to Level 3 and 4. Level 3 and 4 also had warnings about Self-Equilibrium issues, but forces were smaller and don't seem to cause any wild issues like I'm seeing with Level 2.

Before I just "ignore" this by setting these load cases to "Normal" I want to verify the implications and causes. I suspect I have some of these issues because I have intermediate "ramp" floors between the main levels with some "Other material" dummy framing that isn't included in the Concept models. All of my columns and slab areas would be included in the lateral analysis in RAM Concept, but these are all "gravity" members in RSS with all the lateral taken by shearwall cores. And I suspect the problem is compounded at Level 2 because I have a low roof area that is steel framed, and I even have some concrete columns supporting only steel beams that I had to put dummy slab over in Concept just to get the Level 2 model to run.

  1. Most of the discussion I've seen about SE problems center around User Nodal Loads causing issues. To clarify, I don't have any User lateral loads. All my loads are generated from RAM Frame using Equivalent Lateral Force Procedure.
  2. Given the issues with "other" dummies, "gravity" concrete columns in RAM SS, and steel framing combo, will the seismic load ever be in equilibrium in Concept?
  3. Here's an example of what I see in the Calc Log: 

          

Most of the forces are tiny other than the Z forces at node 70 and node 9305 (the southwest and southeast corners of the structure) with shearwalls underneath each of these corners. Can you help me understand why two stabilizing forces added here (that seemingly would be transferred directly to walls below) can affect span designs so significantly even 180+ feet away (not in the same line in either direction, etc)?

4. Do I have any choice but to just choose the "Normal" option for my seismic loadings in Concept and just confirm that the lateral effects on the slab are resolved through other methods?

  • Please share your model using our Secure File Upload and post back here after it is uploaded.



  • The stabilization error is typically caused by the following:

    1) Meshing problems

    2) Members and/or lateral forces associated with those members extend outside the slab edge and are not imported into RAM Concept.

    Both of these issues occur on Level 2.

    Meshing Issues

     1) The modified column and wall elevations between Grid 2' and Grid 3.3' are defined in such a way that there are two different deck slopes (the slope between Grid 2' and Grid 3') is different than the slope between Grid 3' and Grid 3.3'). This results in a meshing problem, because a single deck polygon is modeled over this region and RAM Structural System assumes that the vertices of deck polygons are co-planar. You should either modify the column/wall elevations so that there is a single slope or use 2 separate deck polygons.

    2) The elevation at the top of the column and wall marked below should match.

    3) I suggest using 0" slab edge offsets were you are separating the sloped and flat slabs. A finite element node is placed at all wall ends and slab edge/wall intersections. This creates tightly spaced nodes, which results in bad shell elements, in short wall segments like Wall #56 below. By using a 0" slab edge, the slab edge aligns with the wall end and the interior nodes along the wall segment will be removed.

    Member/Lateral Force Issues

    1) Wall #13 below extends outside the slab edge. Since it is connected to walls that are attached to the diaphragm, it participates in the lateral force resisting system at Level 2. None of the lateral forces are imported since the wall is outside the slab edge. I recommend modeling the slab edge offset from Wall #13 and then using a slab opening inside the wall core.

    2) RAM Concept imports concrete slabs only and not import composite or non-composite decks. The walls marked in the image below are attached to a composite deck. As a result, they are recognized as being outside the concrete slab edge and are not imported into RAM Concept. These walls participate in the lateral system because they are attached to other walls that support the concrete slab and due to deck continuity.

    Here, I would snap the one-way deck to the wall centerline and then add a two-way concrete deck between the wall and slab edge. You can use the effective slab thickness and elastic modulus that you have defined for the one-way deck for that two-way concrete deck, so the properties/stiffness is the same.

    3) I would not be concerned about the gravity columns that are under the one-way deck and are not imported into RAM Concept. Gravity columns are excluded from the RAM Frame Analysis when analyzing the lateral load cases. These columns are not associated with the Lateral Self-Equilibrium errors. RAM Concept will not export reactions for these columns back into RAM Concrete since they are not imported. However, RAM Structural System will automatically use the RAM Concrete Analysis forces to design the columns and the foundations. 

    I have a revised model that I can send you if you would like to see the changes associated with the items above in a working model. If you would like me to send you that model, please private message me and I will send you the model.



  • Thanks for the help Karl.

    I've worked on the mesh issues and think those are resolved... A lot of this stuff always seems to pop up every time I get into ramps because it's always a trial and error to get things modeled such that they work in RAM Frame, RAM Concrete, and Concept. Thanks for helping with those.

    As for the items #2 and 3 you've listed under the "Member/Lateral Force Issues":

    2. You're saying these walls shouldn't come into Concept because but they have always been importing into my Concept model. I always having to add a "dummy" slab thickness on top of them in Concept to get it to run because it didn't like that there was no slab on top. Since I'm doing this (or I could do as you suggested above too), the forces on the walls that Concept brings back into RSS will be incorrect won't they? (This is similar to the issue in the thread below...)

    3. The columns on grid 14' also are coming into Concept. Again, I'm having to model a dummy slab on top of those columns in Concept or it just refuses to analyze the slab. I actually had a separate thread that Seth replied to about this issue here: https://communities.bentley.com/products/ram-staad/f/ram-staad-forum/187278/combining-steel-framing-with-concept

    Seth's solution here would work... since I've got a floor below, I've had the idea of just modeling a steel column at level 2 and then raising up the concrete column height on the story below so the concrete column design height is as close as possible to Level 2.

  • Alan,

    Sorry, you are correct that those walls and column do get imported. There were custom Visible Object settings that I missed that made the walls not appear on the Mesh Input Layer. That was my mistake.

    For the walls, I would use my solution explained above. That will prevent the need for modeling a fake slab when reimporting the model and will have minimum affect on the analysis since the walls are a continuous line of support for the slab.

    For the columns, I would generate the mesh and then delete them on Layers - Element - Standard Plan so that they are removed from the finite element analysis. After doing this, RAM Concept will not calculate a reaction for those columns and reaction will be exported back to RAM Concrete as a result. RAM Concrete will then see no RAM Concept forces for those two columns and automatically use the forces from the RAM Concrete Analysis. The downside to this approach is that you will need to delete the columns on the Element Layer every time the mesh is regenerated.

    I will discuss how such walls and columns are transferred between the programs with the development teams and see if we can improve how they are handled in the future.