Which Method of Load Calculation is Correct?

[kk88818]

I need to resolve a question about the use of load combination and/or load factors in the design of a reinforced concrete footing. Say I need to design both the size of a footing, as well as the design the steel reinforcement, both simple and regular enough. Now, sizing of the footing is done using service loads, while the reinforcement is done using ultimate loads.

For Method 1, in my analysis stage, I am just use the load combinations using allowable stress design to get my service loads, and THEN apply a load factor (1.6, say) to get the ultimate loads for my reinforcement design. In Method 2, I do my analysis using all the possible load combinations using BOTH allowable service design, as well as strength design, or load and reistance factor design, and then using the most critical loads from the ASD combination for my footing design, and then the most critical load overall, or the most critical load from my LRFD load combination for my reinforcement design.

Which method is best, or more accurate? Your thoughts, feedback, suggestions and advice, please.

 

[Lion06]

We typically use ASD combinations for sizing the footing and then use LRFD combinations to design the reinforcement. We typically do this, however, because our program (RAM) is automated and can calc them quickly.
One thing to note for RAM Foundation, it gets very hairy for lateral columns (they provide 0.09 sq in of steel in the top in some instances), and also when As,min controls.

If I were doing it by hand, your approach of factoring the ASD combination by 1.6 doesn't seem unreasonable (even though it is definitely on the conservative side - a good thing).

I might try to sharpen the pencil a little and get a more appropriate factor (maybe 1.4 if your LL is 50% of the total load or 1.5 if the LL is 75% of the total load.

 

[ron9876]

For the condition where lateral loads generate an overturning moment with uplift I haven't ever found a method other than a multiplier on ASD loads since using LRFD can get a completely different soil stress distribution.

 

[kslee1000]

Agree with ron9876, the simple the better.

 

[JStephen]

I think the point of using the LRFD approach is to calculate the completely different soil stress distribution and design accordingly.

For the load cases where you have factored wind plus 0.6 dead load, assuming 1.6 dead load instead is unconservative. This can affect the tie-in to the footing, the rebar required for reversed moment, and in some cases, the maximum moment in the slab.

Where you have partial uplift of the footing, internal forces are no longer proportional to external forces and the external forces should be factored prior to calculating the internal forces.

 

[connect2]

Structural has been Limit states design (LFRD) for a long time here, couple of decades. Last National Building Code 2005 has mandated geotechnical to use only limit states design as well. I used to always carry my specified loads to the ground in anycase for a number of reasons not the least of which was to use the allowable bearing capacity and settlements etc... provided by geotechnical.

 

[abusementpark]

This is why i hate not being able to do ASD for concrete.

 

[ron9876]

If you use LRFD to recalc soil/pile loads you will likely get a different distribution of stresses. For instance if you don't get uplift in a pilecap then you won't need top reinf in the top of the pilecap and the shear stress on the opposite side would be reduced. Don't see how you can use LRFD forces to calc reactions for overturning conditions. Of course a simple column footing is different.

 

[jechols]

You should be able to use LRFD only for sizing the footing and designing the reinforcing. The factor of safety has been replaced with the load factors (even in ASD). For a retaining wall footing the 0.9D+1.6H is approx. the same as 0.6D+H or D+H with a 1.5 S.F.(No longer valid for overturning). To use just LRFD the geotech. must give you ultimate instead of allowable as someone else stated above. Relative stress distributions should be similar. Not saying this is correct just my understanding.

J

 

[ron9876]

If you check something like a pilecap under a shearwall for a 10-20 story building you will see that the end results between using ASD multiplied by an approximate load factor and using LRFD to calc results will be much different.