If we have a series of columns as shown in the picture below and to reduce deflection of the slab between two columns that are 11.7 m apart we placed a drop beam 30 x 100 cm (the slab is 35 cm thick) do we need to check the punching shear for these two columns ? and if we need to check the punching shear do we use the beam thickness (100 cm) or we use the slab thickness (35 cm)

I am asking this particular question because in SAFE for columns C5 through C12 does calculate the punching shear

Punching Shear


1 Answer 1



This can be clearly seen when you consider the load path for everything other than the loads in the region with the beams. Where will a load between C1 and C5 go? Straight to C1 and C5, of course.

A load between C5, C6, C9 and C10 will probably go to the beams and then the columns, so there's no need for punching shear reinforcement in that region, sure, but the loads outside of that area will concentrate right on the columns.

The only question is what the load should be. When calculating the punching shear for column C6, what load should you use?

One option to think about is to just use the load applied in the rectangle bounded by C1-C3-C7-C5 (well, the portion of it that goes to C6). This seems reasonable: this load goes straight to the column (requiring punching shear reinforcement), while the load around C5-C7-C11-C9 first goes to the beams, avoiding the need for punching shear reinforcement. Basically (where full lines represent loads that go straight to the columns and dashed lines go to the beams):

enter image description here

But how sure can you be that some fraction of the load in C5-C7-C11-C9 won't go straight to the column? After all, the column is far more stiff than the beam, so shouldn't it absorb more of the load than'd be trivially expected?

A really conservative calculation would ignore the beams and calculate the punching shear as if all the load went to the columns. Worst case, you're considering around double the real load.

Another, slightly less conservative calculation would use a 2/3 slope to define the load paths that go straight to the column:

enter image description here

If, however, you actually have a FEM program that can accurately calculate the slab-beam-column interaction, there's an easier way to know the punching force: get the column's axial load and then subtract the shear force of the beam at the column. With this, you'll get the total force that was directly absorbed by the column, which is the punching shear force. Just obviously make sure to model the structure accurately.

  • $\begingroup$ Thank you for the representation, So if I Want to be be conservative and let the FEM program (SAFE) to calculate the punching shear itself it is better to remove the beam and see the punching shear for the column (the column will take the load as tributary area); however to be more precise I can find the axial load in column than subtract the shear in the beam and calculate it manually. $\endgroup$
    – J.Daou
    May 2, 2017 at 18:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.