Internal wind pressure

[dozer]

What's the science behind the internal wind pressure as given by ASCE 7? Does the enclosed building classification assume you have some amout of air leakage and that is what is pressurizing the inside of the building or I supposed sucking some air out if the leak is not on the windward side?

In other words, if you had a hermetically sealed box, would you need to consider internal pressure?

 

[btrueblood]

The assumption is that an upwind door can be opened, or an upwind window blown in.

 

[JAE]

if you had a hermetically sealed box .... you'd still have internal pressure.


Internal pressure is the inward or outward force due to differences in the atmospheric pressure between inside and out. Is not directly related to wind movement, which is what the other wind pressures are based upon.

 

[apsix]

In other words, if you had a hermetically sealed box, would you need to consider internal pressure?

For the purposes of engineering design, no, provided it remains sealed.

the inward or outward force due to differences in the atmospheric pressure between inside and out.

I would call this the force due to the net pressure.

 

[dozer]

Maybe I should have said "assuming the gauge pressure inside the box is zero." I'm just talking about due to wind. Anyway apsix answered my question. Thanks.

 

[Tomfh]

In other words, if you had a hermetically sealed box, would you need to consider internal pressure?

No, but you cannot achieve a sealed box. Even with all doors and windows closed it is leaky. Do you imagine if you turned your building upside down and tried to float it that it would float? Of course not, it would leak like a sieve.

 

[haynewp]

I disagree, even if you had a sealed box with gauge pressure=zero in the box, the box would still feel the pressure difference between the inside and the outside.

 

[Tomfh]

If gauge pressure in the box is fixed at zero then by definition there are no internal changes in pressure to worry about.

 

[JedClampett]

By the way, this isn't something you can blame on ASCE 7. UBC also has always had internal wind pressures.

 

[mitchelon]

Folks, I lived in Puerto Rico for 24 years (lots of hurricanes), and I know there was no way to hermetically seal my house. So, during hurricanes, we used to close all windows but those in the “leeward wall.” If we closed all windows/doors, especially those in the leeward wall, we would feel the pressure inside the house trying to blow the windows out. We had an enclosed area in the house with a metal deck roof, which the interior pressure used to lift up. That being said, sometimes it is difficult to comprehend some of the code provisions because it may not really apply to your particular situation, but these things do happen.

 

[miecz]

It seems to me, that if wind blows on a sealed box, the windward face will flex and deflect inward, creating an internal pressure, that in turn, will flex the other surfaces outward, relieveing some of the pressure. If the walls of the box were rigid, an impossibility, then no internal pressure would result.

 

[JAE]

If you had a completely sealed box, then the air inside that box is at a set pressure. As the weather changes, the pressure outside that box changes (you've certainly heard weather reporters talk about "low pressure moving in" etc).

So when you get a net difference between a set internal pressure and a different exterior pressure, you get a net pressure applied to the walls of the box.

This is not necessarily the full internal pressure effect used in ASCE 7. It's factors do adjust depending on the number of openings .... more openings resulting in more internal pressure - my assertion above, I think, is somewhat wrong in that the internal pressure of a "box/building" is more dependent on moving wind, flowing into a partially enclosed space vs. net differences in atmospheric pressure.

 

[civilperson]

Moving air has a lower pressure than stationary air. The gage reading of zero would not reflect the pressure drop in the exterior due to high velocity wind. The pressure drop can be extreme, (760 to 700 in a matter of minutes at the eye of a storm).

 

[UcfSE]

The internal pressure coefficients were derived from wind tunnel tests.

I agree with JAE's line of reasoning.

 

[ron9876]

The internal pressure included in ASCE7 considers leakage/breakage. Note that the coefficient can be + or -. This is because the pressure is positive if breakage on the windward side and negative due to breakage on the leeward side. You are required to consider which is the worst case for the item that you are designing.

 

[dozer]

OK folks, here's where I'm coming from. In the process world you can have some very large structures that look like buildings but are not. No doors, no windows and yes, Tom, they are sealed. However, when you call this an "other structure" per ASCE 7 all you get is the net effect. These structures can get so large I feel its more appropriate to apply pressures like you do for a building with outward normals on the side walls and leeward wall. Problem is once you call this a building then that brings up my question about internal pressure.

After listening to everyone's comments and doing some of my own research I'm convinced that I do not need to consider internal pressure as defined by Figure 6-5 of ASCE 7-05.

Yes, I realize something besides wind may happen to cause a change in internal pressure, but that's a separate load case and we consider this.

 

[structuralaggie]

My environmental loads professor in grad school said to always design as a partially enclosed structure becasue this would be more conservative (of course that is what a professor would say). This gives you the highest internal pressures. He gave an example of a garage door failing and blowing inward and then blowing out the side walls due to the increase in internal pressure. Before the door colapsed, the building was enclosed. If there is no danger of any components failing like this, then I don't think there is a problem treating it as an enclosed structure.

FYI Unless I'm wrong, internal pressures only affect components and cladding loading. The mainforce resisting system should see the same loads regardless if it is enclosed or partially enclosed.

 

[csd72]

As a point of reference, I will tell you how the Australian code works for internal pressures.

In non cyclonic (hurricane) regions the code specifies different wind speed from different directions as the prevailing winds are usually from one or two specific directions.

You can then figure out the internal pressure in either a simplified or a detailed procedure. For the detailed procedure you need to fugure out the size of dominant openings (e.g. roller shutter doors or similar) on each wall and then the effective area of 'leakage' on each wall and the roof (taken as up to .05% of surface area for industrial type buildings and as low as .001% for sealed buildings) .

The effective internal pressure depends on:
1: which wall the 'dominant' opening is on compared to wind direction. i.e. what is the external pressure at this opening.
2. the ratio of the area of the openings on this side compared to the total amount of leakage on all other surfaces.

For example if you had a large opening on the windward side you would get an internal pressure equal to the windward external pressure, if it was smaller you would get some percentage of this. If it was on the sides or leeward you would get negative pressures.

Some companies will actually design roof vents to provide a specified amount of internal pressure to help prevent wind uplift.

As with all of the major wind codes, the Australian code is based on statistical analysis of wind tunnel tests. Actual wind loads fluctuate widely and are hard to predict exactly.

 

[apsix]

"The main force resisting system should see the same loads regardless if it is enclosed or partially enclosed."
This is true up to a point; the net horizontal wind load is the same. However loads on individual members such as columns and rafters of a portal frame are affected, and uplift also varies with internal pressure.

 

[structuralaggie]

right, components and cladding. Roof uplift too, I meant with regards to lateral loads.