Point Load from Column to Slab

[kevinj.888]

Hi All,

I recently did an experiment to understand the point load transfer from column to slab. I have three cases below:

A. Point load applied directly to Slab

B. Actual size of column is modelled, point load applied on top of the column

C. Converted UDL over the area of column

Assuming 0.5x0.5m size column and 5x5m slab area, screenshot below is the result I get.

The interesting finding is that by spreading the point load to UDL (Case C), the bending moment seems to be reduced by more than 50%. I have checked the vertical reaction on the slab edge and all three of them tally. 

My questions:
a. Why there seems to be no difference between Case A and B, where I always thought the load distribution will be automatically following the actual column modelled?

b. Will Case C underestimate the design?

Thank you.

[Rafacascudo]

Several ways of doing this

When using the columm , you can activate "reduction of forces" on parameters tab . check the help file to set this correctly regarding mesh size . Search the forum also ,if needed

You can also create a rigid region with the size of your columm with rigid lings , and take the moments at the edge of this rigid area.

Another option is , for the concentrated load model you can take the moments at the columm face using Panel cuts with the size of the columm width and check the integral value of the cut.

Also check this about this matter Singularities in Finite Element Models: Dealing with Red Spots | COMSOL Blog