Equivalent frame FEM

[ericjacques]

Hello again everyone,
I am performing an equivalent frame analysis on a flat plate slab. I have two different distributed loads on either side of my design frame run. Lets assume that one one side has a load of 10 kPa and on the other, 20 kPa.

What is the most efficient way to combine these different loads to determine my fixed end moments?

Do I take a 50/50 (ie 15 kPa) average of the two loads or do I place more emphasis on the larger of the two(say perhaps, 17.5 kPa, or even 20 kPa)?

Your advice would be greatly appreciate!
Thanks,
-EJ

 

[csd72]

You should input it as 2 separate distributed loads, 10 and 20 . Your FEM will definately allow you to do this.

 

[ericjacques]

csd72, good idea. I will take a look at this carefully.

While we are on the topic, I have a shear wall which runs perpendicular to my frame run. The end of the shear wall is directly in the center of my frame run. Therefore, one half of the transverse width of my design frame is fixed rigidly into a wall, while the other is free.

How is this taken into account in the analysis and distribution of FEM?

I was planning on treating the wall as a deep beam and to determine torsional stiffness and assuming the wall would act as a very stiff column to find flexural stiffness. However, I am not sure how to proceed when determining my span FEM's, COF's and stiffnesses.

 

[asixth]

From your first post, it sounds like you may have a partition running down the column gridline with different design loads on either side of this partition.

When performing an equivalent frame analysis, you are essentially taking a 3-dimensional problem and reducing it to a 2-dimensional frame analysis. In terms of the structural analysis, the computer package will take the udl as 15kPa. It is your responsibility as the designer to distribute this moment to the column strip and middle strips respectivley.

My only knowledge of ACI318 is what I have picked up from using American textbooks, so I will apply AS3600 (Australian).

For a floor system with uniform loading, the distribution of moment to the column strip will be around 80% of the negative moment and 50% of the positive moment.

Now for your situation where you have non-uniform loading, I would distribute a larger percentage of moment to the middle strip with the larger design load (20kPa). If you look at the frame runs either side of this run where the design loads are either a uniform 10kPa or 20kPa, you should get a good indication on what the design moment and reinforcement in your middle strips should be.

Now that I think about, I would just assume the larger design load for this run and not try to get too fancy with the lateral distribution of moments.

As for the shear wall, I would treat this the same way I would treat a column. The wall isn't infinitley rigid and not much moment will be transferred into the wall. Like above, it will significantly alter the lateral distribution of the negative moment over the wall/support, less moment will find its way to the column strip because the middle strip can now span directly onto a support.

Just an aside note, I don't think fixed end moments is the correct terminology and may have even confused csd72 with thinking you are referring to a FE package.

This is a long post and I hope everyone can follow.

 

[ericjacques]

asixth, if I were to treat the wall as a column, the question thus becomes how can I determine how much moment should go into the wall and how much should go into free column and middle strips.

Although inexperienced, my gut tells me this lateral distribution should be related to the stiffness of the two halves of the design run....but I am not exactly sure how to perform this analysis. Any ideas?

 

[asixth]

Correct, load will follow the stiffest load path and the stiffest load path will span column-face to column-face, or column-face to wall for your situation. The middle strip that spans onto the wall will be stiffer than the middle strip which is continuous so more moment should be distibuted to the former.

If I were designing something similar to what you have described, I would distribute 80% of moment to the column strip, 5% to the "free" middle strip and 15% to the middle strip which spans onto the wall.

I haven't performed any torsional stiffness calculations to arrive at these values, they are just values I would use by using my intuition on how I believe the slab will behave.

This is a good question, you won't be able to find any examples of this in textbooks so it will be interesting to hear from other engineers on how they would handle this situation.