roof uplift 1

[broncosfan]

I am designing a building in a hurricane area. The wind speed is about 130mph. The building is a large warehouse with numerous dock doors. These dock doors make it partially enclosed. I get a component and cladding wind uplift of 51.3psf on the joists. Is this force reasonable? I was told by an engineer at a well known joist manufacturer that the highest net joist uplift he has ever seen on a joist was about 25psf. Also, with a 55' bay spacing, this uplift on the roof leads to massive footing sizes to hold down and anchor the columns. Should I be using the load combination 0.6D + W to size the footings for uplift? In addition to the weight of the footing itself, I am using the dead weight of the roof, floor slab, and soil on top of the footing. Can I also somehow account for the friction between the sides of the footing and the adjacent soil? Finally, in sizing the weight of the footings, would internal building pressure component of the wind uplift force "cancel out" because it will "push down" on the slab at the same time it is "pushing up" on the roof? Has anybody done this? Thanks.

 

[DaveAtkins]

The Code allows you to use a tributary area, for each joist, of the span squared divided by three. That may help you.
I would consider using MWFRS load, not C & C loads, for the joists, and I would certainly use MWFRS loads for the footings.
You should use 0.6D + W for sizing the footings.
You can use the shear resistance of the soil above the footing to resist uplift. Bowles deals with this in one of his textbooks.
I don't think the internal pressure "pushing down" will help much, since most of the force will go into soil which is not directly above the footing.

DaveAtkins

 

[csd72]

I have done this procedure a number of times overseas, but not in the US.

I have designed buildings in Cyclonic (Hurricane) regions of similar wind speed where the 6" concrete wall panels had to be tied down to prevent uplift, so the 51.3psf does not sound excessive. Check that you are using appropriate area factors for the area supported by the column.

Are you using the slab surrounding the footing? I have used 10 times the thickness of the ground slab past each side of the footing as additional dead weight.

You can use the internal pressure to help resist this, but only locally over the footing. You will find that it will not make that much of a difference.

0.6D +W is the correct load combination if you are using ASCE7.

Hope this helps.

Personally, I would not rely on friction or cohesion unless I had a geotech. give me some definate values.

Hope this

 

[twinnell]

We have designed steel buildings in high wind regions that have a components and cladding roof uplift load around 50 - 60 psf. Just because someone tells you they have never seen loads that high, doesn't mean they know what they are talking about. We always included "Roof Net Uplift Diagram" on our drawings that showed the net uplift pressure for different parts of the roof. We always used the componemts and cladding load and subtracted 0.6D or whatever the code load combination was for wind. As far as collateral load, I've seen some engineers subtract collateral and I have seen some subtract half of the load. I don't subtract the collateral load. You could have a high wind event during construction prior to the installation of the collateral loads. Collateral being lights, HVAC, etc.

 

[civilperson]

Aircraft hangers at military facilities in hurricane regions have the uplift loading that you describe and higher due to the large door areas. Also, internal walls experience external pressure coefficients and must be designed for wind loads.

 

[UcfSE]

You're loads are perfectly reasonable. Like twinnell pointed out, just because someone else hasn't seen it before doesn't make it or you wrong.

Peronally I wouldn't use MWFRS for joists, but that's another thread. When you're tributary area is over 700 ft[sup]2[/sup] ASCE 7 does allow the use of MWFRS pressures instead of C&C pressures. That will help some with big column loads. Yes, you still use the 0.6D+W combination. I take it a step further and use a minimum dead load for that combination instead of the maximum dead load that I may have overestimated for gravity design.

I would not try to use any friction between the footing and its sides. You have no idea if those will get properly compacted, and you may not be deep enough to develop any friction anyway. I've never tried cancelling the internal pressure for a footing. It's a thought, maybe worth some research.

 

[miecz]

The loads look reasonable to me. You may try to reduce the net uplift by adding some stone or pavers to the roof. In lieu of friction restraint, you can mobilize some soil dead weight outside the footprint. See NAVFAC DM7.2, sheet 7.2-171. Bowles has something on this as well. See the chapter "Bearing Capacity of Foundations with uplift or Tension Forces."

 

[lkjh345]

Your loads seem reasonable to me too.

For what its worth: I have run into the same problem with having to design large, heavy footing as hold down anchors at columns in long span buildings subject to high wind uplift forces.

Sometime in this situation, I opt to use helical soil anchors to provide the neccesary hold down strength. If you have several footings where they could be used, and you gain some economy of scale, they are surprisingly economical to install. In the order of $1,000 to $1,500 an anchor. In certain situations this may be cost competitive with pouring a larger footing just to get the weight.

Just a thought.

 

[Lutfi]

I honestly cannot say that the pressure is reasonable or not. I have seen pressure in that magnitude and I have seen pressures that are higher. It all depends on the various variables. Here is my input based on ASCE 7-02:

1. What code did you use to calculate wind pressures?
2. What wind speed was used in the calculations?
3. Is this pressure in zone 1, 2 or 3 of the roof?
4. What is the influence area of the structural component, joist that was used in the calculations? Influence area need not be smaller than the span times 1/3 of span (L*(L/3)).
5. ASCE permits elements to be designed, just like UcfSE stated, as MWFRS if their tributary areas (not influence areas) are over 700 square feet.
6. Having doors in a wall does not automatically qualify the building as partially enclosed per ASCE 7. See section 6.2 for partially enclosed definitions.
7. For steel joists I would use C&C pressures. See FAQ at the back of “Guide to the use of the wind load provisions of ASCE 7-02”.
8. For the columns and the uplift calculations, I would USE MWFRS pressures. I say this based on simple definition that C&C member is that one that receives wind pressures directly from cladding. For example, roof deck to be designed as C&C pressures. However, it is MWFRS when is being designed as diaphragm.
9. I disagree with the joist engineer’s statement. How can 25 psf net uplift be used as blanket statement?
10. Adding dead load may help as suggested. However, it will be at the expense of larger structural members. This may not be the most economical solution not to mention aesthetics.
11. Make sure that you add uplift bridging per SJI. This is required due to stress reversals.


Regards,
Lutfi

 

[grandcasita]

In hurricane damage, you will see doors that fail, allowing the building to become pressurized internally, with resulting failure of the roof deck, windows, walls or other outlets. The initial door failure increases your internal pressure from 0.18qh to nearly full windward stagnation pressure. You end up with a very large outward force/pressure until something else gives. However, the greatest part of the force to the roof is due to the negative velocity pressure over the top.

Don't count on the downward force due to internal pressure, since the roof uplift does not really rely on internal pressure. If you do, you can only count on the 0.18qh.

If you expect this building to survive a full-on hurricane, you best think about what needs to be done, over and above the code minimums. Component failures will occur at a higher local wind that that considered by the MWFRS numbers. Consider that the reduced loads generated by the MWFRS and large trib area C&C numbers rely on averaging the wind forces over a large area.