Uplift factor of 1.5?? 4

[EngineerofSteel]

I received a review on a simple foundation project. I calculated a perimeter footing and column footings for a pre-fab steel building.

The reviewer (who is a good engineer) is requiring that I increase the metal company's uplift by 1.5. I cannot find any mention of this in the UBC/CBC.

He said it is because:
"note that the 1.5 factor is based on the height to width ratio of the building being greater than .5"

I think this is based on sect. 1621.1:

"The base overturning moment for the entire structure, or for any one of its individual primary lateral-resisting elements, shall not exceed 2/3 of the dead-load-resisting moment."

[That part obviously does not refer to uplift.]

(1621.1 cont.) "For an entire structure with a height to width ratio of .5 or less in the wind direction and a maximum height of 60 feet, the combination of the effects of uplift and overturning may be reduced by one third."

This second part refers to a potential reduction. I postulate he has mentally fabricated a new requirement by jumbling the two parts. I found another thread on eng-tips in which another person believed the same as my reviewer, but no support was found by him or any other engineer. see it here:

http://www.eng-tips.com/viewthread.cfm?qid=75910

CAN ANYONE JUSTIFY THE REVIEWER?

Thanks, DD

 

[sundale]

I think your reviewer is perhaps correct if the project is under the UBC. The RATIO (i.e. safety factor) of wind overturning moment to dead load restoring moment needs to be at least 1.5.

'97 UBC had D+W as a load combination (Eqn 12-9) and prescribed a SF=1.5 against overturning (UBC 1621.1)

'03 IBC has 0.6D+W as a basic load combination (Eqn 16-11). IBC 1609.1.3 only requires a SF of 1.5 for the alternate load combinations, which are more or less UBC load combinations.

Note that the reciprocal of 0.6 is 1.67, which makes the IBC more stringent than the UBC as far as having enough concrete as a "deadman" to resist net tension in a metal building column. Many metal building footings that I design are governed by tensile uplift instead of gravity soil pressures.

 

[EngineerofSteel]

sundale,

Your answer highlites my contention to this requirement fromt the reviewer: the 1.5 in 1621.1 applies to overturning, not uplift. The metal building is designed with pinned footings, resulting in only vertical and horizontal design loads.

uplift is not mentioned until the second paragraph.

I do not believe uplift should be multiplied by 1.5.

I am hoping for confirmation with a good back-up reference (in addition to 1621.1) OR a good reasoning why 1.5 is required - not based on 1621.1 ... my reasoning does not see the 1.5 applying to uplift.

Thanks, sundale, I appreciate your input. I agree that if I needed to design to IBC, I would need a SF, but we are still under '97 UBC here in Stanislaus Cty, Calif. However, as you noted, Eqn (12-9) is D + W and the SF applies to overturning, not uplift.

 

[UcfSE]

The uplift you receive from the PEMB manufacturer may well be due to overturning. Did they break down the uplift into uplift due to wind tributary to the roof alone and uplift due to overturning of the building frames?

 

[JStephen]

It's confusing to me as well- but largely because the structures I design don't have design uplift (IE, vertical pressure up on the roof), but only design overturning- so in my mind, overturning=uplift. Maybe he's doing the same thing. And of course, if it's due to wind, they'd always occur together anyway.

Something you might look at- if there's a load case or safety factor for overturning, then there's bound to be a similar case for uplift- and if you can't find it, perhaps the 2/3 requirement is to be applied for the combination, not overturning only.

 

[EngineerofSteel]

UcfSE,

The load reactions are listed only by load "Id", such as DL+CL+LL+Windright, the reactions are then subheaded with two categories only: vert & horz

Hmmm...

what they DO do is separate "suction" from lateral wind loads...

what are you thinking? Prorate the force due to overturning?

Thanks, DD

 

[EngineerofSteel]

JStephen,

You may be right. The idea that uplift defined as part of overturning is new to me.

I still don't like it... If the designer produced only vertical and horizontal loads at the base, it is because the base is a pinned connection and the haunches are stiff. If the haunches and vertical members have already been designed to keep the building from collapsing, I hate to weight down my client's pocketbook with a 10 additional cubic yards of concrete.

Thanks, DD

 

[haynewp]

I don't have the codes with me right now.

But you should have a factor of safety in with the pure uplift of a building the same way you would overturning.
You must apply a 1.5 or whatever safety factor for overturning. It is the same reasoning for pure uplift. If it is a typical gable rigid frame, then you may have uplift forces being derived from uplift and overturning on the rigid body (perpendicular to ridge). But the values will be without any safety factors if unfactored wind was applied when the values were generated by their software.

I think you should apply the 1.5 safety factor to the uplift value, whether it be from overturning or pure uplift, or both. But don't do it twice (reduce the dead load and multiply uplift by 1.5).

 

[Lutfi]

I would use SF 1.5 against uplift and over turning. I have done so my entire career! I think it is the right way.

When evaluating over turning effects on footings and if have uplift on the footing, make sure you reduce the footing weight by the uplift force!!





Regards,
Lutfi

 

[sundale]

I think that the metal building manufacturer has already done the overturning statics for you. Look at the load cases they give you, and see where the net tension load is coming from. It is likely a lateral load condition, not a pure uplift condition. Of course, most PEMB calc's are Byzantine gobblygook to me...

If there is a column shear combined with a net tension, then the overturning moment has been "distilled" to a net tension FORCE demand at the anchor bolts. Your volume of concrete in the footing, slab, pilasters, grade beams, etc, and any dirt over the footing, should weigh 150% of this tension demand.

I agree with Lufti; any net tension should have at least a SF=1.5 in regard to a deadload resisting force/moment.

 

[EngineerofSteel]

Thank you all. I will go with this unanimous decision.

However, I still think that straight uplift is obviously countered by straight and equal deadweight.

In designing a footing and connections, it makes sense to me that there should be a SF of 1.5 (+importance SF, if any). There are too many variable to get an accurate detailing... soft spots in soil (that the contractor will disregard), contractor shortcuts & substitutions, poor measurements, connections, etcetera. Pure uplift, thought... at least theoretically, 1:1 + a small rock = superiority.

However, I think the true reason that uplift requires a 1.5 is this: theory (pinned connection theory) does not ideally equal reality, and some (at least small amount) of moment will come into the column. Therefore, even a "pure" or "distilled" vert & horiz resultant vector system is not ture & accurate.... and the moment resisting factor of 1.5 should be used.

Agree? Disagree? I need some reasoning to back this up with the BOSS.

Thanks, DD

 

[haynewp]

I think there should be a safety factor in the uplift just like there is a safety factor in everything else on the building, regardless of the actual column end condition.

 

[EngineerofSteel]

Of course there is a safety factor in the calculation of every member within the building, in the design wind load, in the UBC equations, in the allowable soil bearing pressure, etcetera.

Adding ANOTHER SF on top of all the already given increases does not make me "think" it is justified. These buildings don't often blow over. They usually fail from snow loads.

-DD

 

[haynewp]

If you are using ASD, then there may not be that much of a safety factor in the wind load you develop. It doesn't make sense to me to have all the safety factors in the steel frame and its connections, then you get down to the footing and just use a 1:1 ratio.

 

[akastud]

Dairy Designer,

I also use the 1.5 FS for the uplift load, but I always include the superimposed soil weight on the footing, and I have even assumed that the footing would extracta wedge of soil that extends from the top perimeter of the footing and goes out at 45 degrees for additional dead load to resist this force.

akastud

 

[EngineerofSteel]

haynewp,

Thanks for a clear insight into what I am saying. You are right, I am saying: "If the steel, bracing, connections and etcetera are designed to 150%, and I design the footing at a 1:1 with the aforementioned design, then I have designed a footing at 1.5:1 of the actual, expected forces.
On the other hand, if I take the design loads (already at 1.5:1) and introduce another FS of 1.5 in sizing the footing, then my footing becomes 2.25:1
You are saying, "why lose the insurance?" I am saying, I am not losing the insurance, but overpaying on my premium. In other words: Why should I design a footing far stronger than the building it supports?
I researched and found that PEMBs tend to fail under snow load. I did not find any mention of the building "sliding" or the foundation lifting up or the foundation splitting as the wall failed due to a moment force. Does anyone know of any?
I had not looked at it in this way. You have clarified it very well, THANKS!

-DD

 

[EngineerofSteel]

akastud,

I have seen addition of a soil cone before, and I read a thread on foundation engineering for tanks in this forum which gave case examples supporting this (footings coming up with cones of earth wedged between and alongside).

So, does the cone apply for download AND uplift. since you say "from the top perimeter of the footing" I think you are talking about download. For uplift, I expect the same rule applies.

I did add a force for the soil. I calculate a friction force using the passive soil pressure times the friction coefficient in Table 18-1-A

My reasoning is this: sliding friction results from pressure between the bottom of the footing and the soil below. This is the 18-1-A coefficient * the DL. I have the same effect against uplift: contact between concrete and soil and pressure. I calculate the pressure as 0 for the 1st foot of depth, 300 for the 2nd, 450 for the third, etc... I multiply this by the square footage of contact area.

Agree? Disagree?

 

[UcfSE]

DD, are you saying you are using soil friction to resist uplift? I think I didn't read your post very clearly.

When you design a footing 1:1 to meet load capacity of the structure that has a FS of 1.5, then your footing will also have a FS of 1.5. Otherwise if you design your footing to be 1:1 compared to the loading, then obviously you have a FS of 1 that isn't acceptable.

We don't want to over over do it by compounding safety factors, but we also don't want to let our simplifying assumptions made during the design process to undermine our safety factor either. For instance, if you assumed 20psf dead load for your structure for gravity loading when your actual is somewhere between, say, 14psf and 18psf, then you are a little conservative. If you are using that same assumed dead load to calculate net uplift, then you are being unconservative. How confident are you that the contractor will provide at least as much concrete as you specify? Will the soil above the footing be compacted to the same unit weight as that below the footing or is it just loose fill? Is that soil even there?

In my opinion it's not fair to say that most PEMB fail under snow load. If they had been designed properly they wouldn't have failed in the first place, so how can you make a comparison to the footing design when the footing wasn't even tested by a design load event? Properly designing the structure will allow the lateral systems to get the load to the footing, and then the footing better be done properly!

 

[EngineerofSteel]

I am saying I am using soil friction to resist uplift. Here is my laboratory experiment: I buried a brick and placed another brick on top of the earth. Then I lifted each one and noticed that the buried brick was very tough to lift, while the brick on dirt was easily lifted. I attributed the difference to burial. Perhaps there is some other way to calc this? Is there a reason this is disallowed? The friction I am using is the friction against the vertical sides, not the friction against the base of the footing.

I did not mean to convey that Most PEMBs fail under snow load. I meant to say that all of the articles I found in my Google search about PEMB failure were about failure due to snow load.

Of course the footings weren't tested by a design load event: the building failed before the footing! That is my point: Why install 2 CUYD of concrete when the building will fail without testing even 1/2 CUYD? Exactly.

I understand about the two-way use of DL in calcing against uplift and for DL. Thanks.

Thanks, DD

 

[UcfSE]

I've never heard of using soil friction to resist uplift the way you have described, but that's not to say it isn't there. I do think you should have more quantifiable data than a buried brick, and should use a brick large enough to be a footing, before you use it in design, imho. You may have some suction at the bottom of the brick also, particularly in moist clayey soil. How much of the added resistance you found do you think could be attributed solely to the soil weight? We're also back to the compaction and soil preparation that takes place after the footings are cast, which is probably little if any. You won't have much friction in a loosely compacted soil of low unit weight as you know. Of course, if you have all of that data already, more power to you!

What kind of forces where you getting for each brick, out of curiosity? What was the depth of the brick to width ratio, D/B and D/L? With most shallow footings, you're looking at burying a brick a few inches in order to have a similar depth to width ratio as a typical shallow footing. It doesn't seem like a few inches would help much beyond the added soil weight.

My point with the failure at snow load was that the comparison is invalid because you observed an improperly designed structure. Who's to say though? I do understand your point and I wasn't there, but I don't buy into that reasoning personally.