Punching Shear on a Footing

[JoshPlumSE]

Take an example of a simple spread footing subjected to an axial load Pu applied through a pedestal. Assume that there isn't any applied moment, though the axial load may be considered to be eccentric. There is also some soil overburden pressure and the self weight of the footing itself.

When you calculate the required punching shear strength, you start with the applied load Pu. But, then you get to reduce it a bit because of the soil pressure below which is counter-acting the applied load, correct?

Obviously, you can only use the amount of force / soil pressure that is within the assumed failure perimeter. This is, after all, the only soil pressure that directly opposes the load Pu.

The question is how do you calculate the force reduction in the punching shear demand? What pressure do you use? Do you base it on the TOTAL applied load (which includes the effects of fooing self weight and the soil overburden)? Or, do you ONLY subtract out the portion of the pressure that directly comes from the applied pedestal load?

I know how I would do this, but I wanted to get other folks opinions on the subject.


Thanks in advance for any input,

Josh

 

[hokie66]

I don't do any deduction for the area under the column.

 

[structuresguy]

I would not reduce the punching shear at all. If you are that close that it makes a difference, you have underdesigned your foundation, IMO. Punching shear, whether in footings or slabs, is a critical failure mode (due to lack of ductility and real possibility of collapse) that I would never be that close on punching shear strength. My recommendation is to take your minimum footing thickness and increase it by 25%.

 

[asixth]

Strictly speaking you can reduce the punching shear force to account for the column load that will transfer straight to the soil, you are only looking at the shear stress at the critical perimeter, however most software packages will simply use the column reaction as your punching shear force. I also do this and for your example, if punching shear is critical I would simply thicken the spread footing.

 

[JoshPlumSE]

This is actually a theoretical discussion that I got into with another engineer. It does not involve an actual foundation. It's really more of question of "can you think of any reasons why this would NOT be allowed by code?" Or, "Is there a solid theoretical basis for not allowing this reduction?"

My opinion has been that you can reduce the punching shear demand by WHATEVER soil pressure occurs directly underneath the punching shear perimeter. And, that it doesn't matter whether this soil pressure was developed as a result of the pedestal load or the footing self weight or the soil overburden.

If your footing/mat actually experiences this soil pressure then you can use it to reduce the demand. I cannot see anything in the code that would specifically forbid me from being allowed to do this.

Whether or not we feel comfortable as engineers with the reduction in our safety factors when we do this is another question.

Josh

 

[Lion06]

I'm not sure I'm following the question exactly, but I don't typically use Pu as my punching shear value. I draw out the footing and show the bo perimeter (pedestal +d/2 all the way around). I work out the soil pressures and come up with a Vu - which is the force that is outside of the bo perimeter.

 

[Lion06]

Also, I would not include the footing self weight or soil overburden as they act on the whole footing. If you have a ftg with no load on it, but buried 20 ft down you wouldn't be worried about punching shear even though you have about 22ksf on it.

 

[seny]

soil will be checked for service, design of concrete for ultimate. in real life you never have soil pressure equal to ultimate column reaction. so you can simply check punching for it based on appropriate perimeter and it will be enough conservative.

 

[JoshPlumSE]

Structural EIT -

Your method of using the soil pressure that occurs outside the punching shear perimeter results in the same thing as mine.... I'm just looking at it upside down (or you are).

But, to point out a minor difference. The soil pressure due to overburden will have a total magnitude that cancels out the applied over burden. However, that force can be spread out over a different area (because of a triangular soil bearing profile or becasue the area of the pedestal). In most cases the differences between the two methods should be very small.

Like I said earlier, this is a fairly theoretical discussion. But, I work for a software company. Therefore, we try to take into account every minute little detail. Even if my hand calculations wouldn't normally account for this force reduction, I think it is the more theoretically correct way to calculate the demand.

 

[hokie66]

I understand SEIT's approach, but disagree. As others have pointed out, punching shear calculations should be very conservative. Code punching shear provisions are not intended to model the actual behaviour, but rather provide a safe approach.

JoshPlum, I would not make a deduction in your model. I believe most engineers, in checking your software, would believe you are in error and would then distrust your results.

 

[JoshPlumSE]

I don't think anyone is disagreeing with the theory of reducing the punching shear based on the counter-acting soil bearing pressure.

Instead, the prevailing argument seems to be "I'm not comfortable that the code provisions can assure an adequal level of safety for punching shear. Therefore, I try to be make sure that my calculations are a bit mor conservative."

I'm not poking fun at that argument. If my mat foundation were at 98% of punching shear capacity, I would personally increase the thickness of the mat... at least if I were to seal the drawings. I just want to make sure that no one has a solid argument against the technique...one either based on code provisions, first principles, or with a solid basis in theory.

 

[bootlegend]

I've always done it the way Structural EIT and JoshPlum are suggesting. I think the PCA notes example do it this way also.

 

[miecz]

PCA Notes Ex 22.2 calculates the shear as SEIT outlined. Ex. 22.1 and 22.2 do not deduct the weight of the concrete or the soil overburden, as these don't contribute to the shear in the first place. That is, P[sub]u[/sub] does not include the soil overburden or the weight of the concrete, or the live load on the overburden.

 

[KBVT]

I always picture the punching shear Vu value as the amount of force that must be transferred from the applied load across the failure plane to the remainder of the footing. Hence, the vertical applied soil load and concrete self-weight are not transferred across the failure plane as they are uniform loads applied to the entire footing. Only axial force and moment applied within the punching shear perimeter should be considered.

 

[Zafer1234]

I am providing a punching strength calculation that can be used for designing two way beamless slabs. The calculation also takes the moments existing at the joint into consideration. The calculation was carried out according to Russian Standarts (SNiP). The article can be downloaded from

http://www.structoris.co.uk/index.php?aid=91.

I hope it will be useful.

Zafer Geçim
Civil Engineer, M.Sc.
Structoris Engineering Consulting Contracting Co.

 

[jike]

For problems where the eccentricity is within the kern area, I calculate the punching shear as SEIT has indicated without considering overburden or self weight.

However, when the eccentricity is beyond the kern area and I need the footing self weight and overburden to help resist overturning, then I use those forces to calculate soil pressure and thus punching shear.

 

[JoshPlumSE]

Jike -

There are many cases where when you do not include the SW and OB then the eccentricity will be beyond the kern. But, when you do include the SW and OB then the revised eccentricity will be within the kern.

I think this method still makes the most sense for hand calculations, but it is based on some engineering judgment calls about when you use one method or the other.

Therefore, it ends up being easier (at least for a program) to treat all cases the same rather than to try and switch back and forth based on a semi-arbitrary criteria.

 

[Ltdog]

The pressure is the net reaction for unit area of ground soil corresponding to the fundamental combination of load effects,of which after the self-weight of foundation and the weight of upper applied soils are reducted.

 

[miecz]

JoshPlum

For programming purposes, simply include the SW and OB, and the resulting increase in soil pressure.

For footings carrying axial load only, with no moment, the SW and OB result in increased calculated soil pressure, with no net change in punching shear.

It would seem that a footing carrying a moment in addition to axial force would fail sooner in punching shear, as the punching shear stress would be higher on the side with the increased soil pressure. I'm not sure if this is addressed anywhere in the code, and I don't remember seeing a published solution.

 

[jike]

JoshPlum:

I agree, (for programming) it is easier and more consistent to include SW and OB for all cases. Just make sure that you deduct the resulting soil pressure so that it's affect zeros out for the axial only case.