Footing Uplift 5

[slickdeals]

Folks,
When you calculate the uplift load on a footing, the code requires the use 0.6D + W. Assume this gives a net uplift of 30 kips.

When you calculate your resistance to uplift in terms of the footing weight and the weight of a truncated soil pyramid (based on a 30 degree angle), do you use a 0.6 factor on the uplift resistance and compare it to the above calculated uplift of 30 kips?

It seems like double dipping (very conservative) to use a 0.6 factor on the resistance side also. Thoughts?

 

[BAretired]

abusementpark,

If I have offended you, I sincerely apologize. It was not my intention to do so.

The Canadian code does not recognize the 0.6DL + W criteria because ASD (Allowable Strength Design) is no longer an acceptable standard in Canada. The standard we use is LSD (Limit States Design) which corresponds to your LRFD. It has been about thirty years since I used ASD and I may be a little rusty so please bear with me.

The equivalent load combination in LSD is (1.25D or 0.9D) +1.5L. When dead load contributes to the load under consideration, it must be taken as 1.25D to provide a safety factor against collapse. When dead load acts in opposition to the load being considered, it must be taken as 0.9D to account for the possibility that it may have been overestimated. Live load must be taken as 1.5L when it contributes to the load and zero when it acts in opposition because live load is transitory.

Basically, anytime you have uplift being reduced due to the 0.6D factor on something that is guaranteed to be there (i.e. footing, pile cap, concrete shear wall, etc.) along with a geotechnical factor of safety on the uplift resistance of a foundation element (i.e. friction piles, drilled shafts, adhesion on footings).

If there is a net uplift on a pile, there must be sufficient reinforcement between the column and pile to safely resist the uplift. Using ASD, that would mean designing the reinforcement for an allowable stress of about 60% of it's yield. Using LSD, we would take the factored uplift and design the steel for 90% of yield. Either way, we would arrive at the same area of steel.

Now, column uplift has been delivered to the foundation. If the foundation weighs exactly as much as the uplift, then theoretically everything is balanced, but there is no safety factor against possible increase in uplift. This is not in keeping with the principles of structural design. A safety factor of 1.5 is needed. This means that the foundation must weigh 0.67W where W is the maximum uplift force expected. To allow for possible differences in unit weight of concrete, volume of concrete, etc. your code uses the figure of 0.6W.

The dead weight of the foundation contributes 0.6D to resisting uplift using ASD or 0.9D using LSD.

The remainder of the required uplift must be made up using skin friction of the pile or adhesion of the footing with an appropriate safety factor, as determined by the geotechnical engineer.

BA

 

[Joepaterno]

I'm one of those younger structural engineers that is now confused because of this thread.

I believe I have an example that perhaps will spur some more interesting debate. I have a stack, 150' tall, 32' in diameter, on a massive, rigid pile-cap foundation. The top of concrete is above grade, and no embedment effects are considered on the edges of the pile cap. I've been supplied with vendor loads for the stack as a whole. I've completed the design, and believe it to be adequate, but I think this provides a good example for the discussion.

Here are my design parameters:

Vendor Supplied Stack DL: 337 kip
Total Foundation DL: 1005 kip
Wind Load Overturning Moment: 7663 kip*ft
Bouyant Force on Bottom of Concrete: 380 kip

Using a STAAD model, I've identified that the tension in my piles is the governing failure mode. Using the load combination 0.6x(DL stack) + 1.0x(DL Pile Cap) + 1.0xWL + 1.0xH (BOUYANT FORCE)

Note that I took the full load of the mat to resist the overturning. I acheive a pile tension of 13.3 kips. My allowable pile tension, as reported in the geotech report is 13.5kips. This allowable load already includes a 2.5 geotechnical capacity saftey factor.

Considering both a) the foundation DL is known with a high level of accuracy (our constuction tolerance make the design volume of concrete a minimum) and b) the geotechnical allowable capacity of the pile includes a factor of safety > 2.0, is this design acceptable?

I argue that this design is valid, in no case can the wind load from the stack be applied to the foundation with less than the full dead load of the mat being there. I used 0.6 for the DL of the stack, as it is conceivable that the stack liner could be taken out during maintenance. Additionally, the 1.5 factor of safety for foundation overturning the 0.6 factor is supposed to replace does not apply, as my foundation is pile supported.

Note that I have a tight site, with no chance of expanding the foundation plan area and spread out the pile group. I can't see increasing an already 5'3" thick mat, costing the client tens of thousands of dollars, as the intent of the code, in using only 18% of the foundation weight to when calcuating the tension in the piles due to wind. (150 pcf * 0.6 - 62.4pcf = 27.6pcf, 27.6/150 = 18%)

Anyone agree / have material I can use to prove that the codes use of 0.6 for the DL of mat is absolutely necessary?

 

[BAretired]

Joepaterno,

In my opinion:

DL = 337 + 1005 - 380 = 962k
0.6*DL = 577k

If N = number of piles, then each pile has an allowable resistance of (13.5 + 577/N) kips.

BA

 

[csd72]

This may have been covered but:

Have you included the weight of the slab on ground in your resistance calculations? we used to use a distance of ten times the thickness of the slab in each direction.

Also for a very small added benefit, your internal pressure is also pushing down on the ground as well as up (but yes it is very small.

I have also seen people use a nominal amount of friction around the sides of the footing.

 

[Lion06]

Joe-

I think you still have to use 0.6 times your pile cap weight. It's a DL and the code doesn't allow you to decide which DL is overblown and which is on the money. That's the way I rerad it anyway. When in doubt I always err on the conservative side - it helps me sleep better.

 

[steellion]

Joe, I'd use 0.6 x Dead. It's been debated on this thread, but I feel that the 0.6 factor has much less to do with the lack of knowledge of the true Dead Load as it does with providing an adequate safety factor. Remember, while the Dead Load is well-known, the Wind Load and Buoyancy Force are highly variable.

The factor of safety for the piles are for individual piles, backfigured from the yield point. I think you would still need the factor of safety against overturning that the 0.6D+1.0W+1.0H provides.

 

[structuresguy]

Yep, I'm with Steellion and EIT, you have to include the dead load of the pile cap with the dead load of the stack, and multiply it all by 0.6 factor. You don't get to decide which dead loads you want to apply this factor to. Not when the code sl clearly mandates the load combination. TO do otherwise opens you up to significant risk, should anything ever happen to your structure. If you did not follow the code, you would surely lose at trial.

 

[Joepaterno]

Thanks for your opinions, but I'm not sure I agree.

Using the logic that the 0.6 D was intended to replace the 1.5 geotechnical factor of safety for mat foundations for uplift, which is no longer used, why should the pile geotechnical safety factor for tension be used? (Keep reading, I know how that sounds)

If this was a mat foundation, I would simply use the 0.6D, and check against the tension with no factor of safety. But it would be acceptable conversely in previous codes to use 1.0 (or 0.9 D) and account for the factor of safety at the end. Similarly, the factor of safety my design counts on is embedded with the pile geotechnical allowable load. This allowable load has a 3.0 factor of safety. I don't think this allowable load should be checked against service level loads that have that inherent safety factor. It would be equivalent to taking a factor of safety to the applied load, and a factor of safety to the resistance. Total factor of safety would be 1 / 0.6 * 3 = 5, which seems very high.

Additionally, I've made the following conservative assumptions:

1) Water Table is actually at bottom of foundation, the full bouyant load on foundation is taken to account for extremely unlikely event of the water table being top of ground surface.

2) No cohesion taken on sides of 5'3" thick mat in cohesive soil.

Did I change anyone's mind?

 

[Lion06]

Not yet. There's more than the geitech safety factor to consider. Even if I were to agree with your approach, you still need to the 0.6 on the cap for the connection of the pile to cap, for the cap itself, and for the actual pile capacity. The only thing your approah gets you out of is the required rock socket length (or whatever the transfer mechanism may be from pile to soil/rock). Everything above that still needs the 0.6 factor on the cap self weight - even usin your approach.

If you have a footing with a large overturning moment (from wind) such that the resultant axial load is close to the edge of the footing (but still within the footprint of the footing) and using a 0.6 factor on the footing self weight results in the soil bearing pressure being too high would you still take the same stance and say you can use the full footing weight without the 0.6 factor since the soil bearing has a safety factor on it already? I wouldn't and I don't think that's the intent of the code.

 

[Joepaterno]

Also, in the California Building Code, the alternate allowable load combinations do permit the use of 0.9D per Section 1605.3.2

 

[BAretired]

Vendor Supplied Stack DL: 337 kip
Total Foundation DL: 1005 kip
Wind Load Overturning Moment: 7663 kip*ft
Buoyant Force on Bottom of Concrete: 380 kip

The dead load is weight of Stack + Foundation - Buoyant force: i.e. DL = 337 + 1005 -380 = 962 kips. So 0.6*DL = 577 kips.

The soil report provides an allowable tension of 13.5 kips based presumably on skin friction. That cannot be changed by anyone but the soils engineers.

If there are n piles uniformly spaced around the periphery of the pile cap, each pile is capable of resisting an allowable tension of (13.5 + 577/n) kips.

The wind will stress some piles in tension, others in compression. The greatest uplift will occur to one or two piles in line with the wind direction and the center of pad. For those one or two piles, the calculated wind uplift should not exceed the value given above.

Does the foundation satisfy that requirement?

BA

 

[csd72]

load factors are for variability in loads, material factors are for variability in materials.

The minute we start to think that things will be built/behave exactly as we design them is the minute that we start to get failures.

 

[BAretired]

Joepaterno says:

If this was a mat foundation, I would simply use the 0.6D, and check against the tension with no factor of safety. But it would be acceptable conversely in previous codes to use 1.0 (or 0.9 D) and account for the factor of safety at the end. Similarly, the factor of safety my design counts on is embedded with the pile geotechnical allowable load. This allowable load has a 3.0 factor of safety. I don't think this allowable load should be checked against service level loads that have that inherent safety factor. It would be equivalent to taking a factor of safety to the applied load, and a factor of safety to the resistance. Total factor of safety would be 1 / 0.6 * 3 = 5, which seems very high.

This is faulty logic. Resistance to the applied wind load is resisted by two separate entities, dead load and skin friction, each taking part of the applied load. For a given pile:

0.6*DL + Allowable Friction = W(service load)

The allowable friction is determined by geotechnical considerations. In this case, it is 13.5 kips/pile. The factor of safety for soil friction may be expected to be substantially higher than the assessment of dead load because soil properties are highly variable. It may be as much as 2.5 or 3.0 but it applies only to the term "Allowable Friction", not to DL. Combining the two factors to form a total factor of safety of 5 is absurd.

BA