Effects of global geometric imperfections when checking foundation allowable bearing stress. 1

[hetgen]

Is it too conservative to consider notional loads (horizontal forces added to the structure to account for the effects of global geometric imperfections) when checking foundation allowable bearing stress? The global imperfections are not transient load they are there permanently.

But we normally apply a factor of safety of 3 against the ultimate bearing capacity to arrive at the allowable bearing capacity, will that safety factor somehow account for it?

Many thanks.

 

[JoshPlumSE]

It certainly could be reasonable. If you know what the imperfections are, then there is no reason not to include them.

However, most of the time, you don't know what the actual imperfections are and you're just guessing in order to capture the worst case scenario. Also, when you're guessing, there is a good chance the actual imperfections tend to cancel each other out at the foundation level.

It's pretty rare for those imperfections to have a major effect on design. You really have to have a pretty slender structure for those imperfections to start causing major design issues.

 

[KootK]

I'd like to hear more about your process. For the most part, notional loads are meant to represent "perturbations" or "instability instigators" rather than discrete loads in their own right. Which is not to say that the construction misalignments etc are not real loads, they are So I see the process as generally being:

1) Analyze the superstructure with the notional loads to get overturning forces etc that are properly amplified for second order effects.

2) Use the amplified forces from #1 to design your foundations. In this sense, the foundation design does indeed incorporate the notional loads. In this context, I would expect notional loads to have the greatest impact on overturning induced uplift loads.

Is that not your process here hetgen?

Alternately, sometimes I design foundations by hand long before I do the detailed superstructure design. In those scenarios, I wouldn't normally bother with notional loads as I'd expect them to get drowned out in the noise of all the other, more significant stuff. That said, one does always have to keep their wits about them and I've no doubt that, once in a while, a situation may indeed arise where foundations should be designed giving explicit consideration to notional loads.

I'm leery of saying that factors of safety cover notional loads. To my knowledge, factors of safety are not generally meant to cover analysis inaccuracies.

 

[hetgen]

@Kootk, I consider notional loads on all combinations, including service load combinations.

I'm doing a peer review and the designer said the notional loads should only be considered for ultimate strength design. He said it is too conservative to consider allowable strength soil factor and ultimate strength design stability factor at the same time.

I responded and said, the structure imperfection (i.e, the national load) is not a factor it there permanently. He seems to think the allowable stress factor on the soil strength side should account for it.

 

[JoshPlumSE]

The global imperfections are not transient load they are there permanently.

The ACTUAL imperfections are permanent, but chances are you have very little idea of what they are. Hence, you are using notional loads or an assumed imperfection as the "instability instigator" as KootK suggested (I like that term).

My previous response is mostly talking about using notional loads in combination with gravity only loading. For structures that have very little in the way of lateral loads, this may be a good thing to look at for foundation design. Like, foundations for internal mezzanines in a low seismic area.

But, I would think that, for any structure with a normal wind or seismic load, the notional load case wouldn't really control the foundation design because the lateral loading is so minor. I've had cases (hydrostatic test loading) where the structure is loaded VERY heavily very briefly to test equipment. In that case, the gravity TEST load is much higher than the normal operating gravity load. It's a good idea to combine that test load with notional loads as it might control the design since seismic and wind forces would be applied to the structure when the gravity loads are much, much lower.

 

[hetgen]

Thanks, Josh.

I agree with you that we don't really know the exact global imperfection, but I have used and have seen others use a figure ranging from H/500 to H/300 depending on the design code and/or project specification.

As I understand it, the national load is not only applied to gravity load combinations as you indirectly implied, it has to be considered over and above the lateral loads. So it is always there with whatever load combination turns out to control the design.
For example, if you are considering the wind load on the +X-axis direction you also have to assume the building to be inclined toward the +X-axis (i.e, you apply the national loads in the +X axis direction over and above the wind load).

My question is do you consider the building tilt ( H/500 or whatever the agreed global imperfection) when you check foundation settlement?

 

[canwesteng]

The tilt should be considered when designing foundations, but the fictional base shear need not be considered. Maybe that is what your designer is getting confused. You could remove the magnitude of your notional loads from the base shear on the foundations if you wanted to. I don't really see how much that can possibly help though, if 0.2% (0.5% in Canada) of gravity as shear is failing the foundation I wouldn't be comfortable with the design anyway.

 

[JoshPlumSE]

As I understand it, the national load is not only applied to gravity load combinations as you indirectly implied, it has to be considered over and above the lateral loads. So it is always there with whatever load combination turns out to control the design.

It depends on your design code for sure. But, if we're talking about the US codes (i.e. AISC) then the notional loads are NOT usually applied in combination with lateral loads. See section 2b (Use of Notional Loads to Represent Imperfections), take a look at item (d):

For Structures in which the ration of maximum 2nd order drift to maximum 1st order drift in all stories is equal to or less than 1.7, it is permissible to apply the notional load only in gravity-only load combinations and not in combinations that include other lateral loads.

In my experience, a structure has to have some real slenderness / drift issues to reach that 1.7 threshold.

 

[hetgen]

Just an update.

According to the Eurocode, I was wrong.

Eurocode is the design code for the building in question, the code explicitly states that the geometric imperfections need not be considered for serviceability limit states.

@JoshPlumSE, Thanks.