Wind Suction Overturning Moment for Container Foundations

[OffshoreWindStructures]

Hi,

I wanted to get some outside opinions on how much suction forces on roofs due to wind contributes to OTM for small modules or shipping containers,

Wind load on containers produces pressure on the windward face and suction on the leeward face and roof. There's a discussion on whether or not to include the roof suction loads (pink in sketch) in the total overturning moment.

I've sketched a picture of the forces and points of application as i understand it. Others in the office don't think it's worthwhile taking the roof suction loads (in pink) into consideration when considering overturning moments from wind and righting moments from self weight about point B.

An additional question is on net coefficients for overall non-cladding loads for foundation checks such as this. Uk/Eurocode guidance suggests a net lateral wind coefficient of 1.2 to 1.3 (i.e. 0.8 for windward pressure and 0.5 leeward suction but does not provide a net overall coefficient for roofs.
Do you use overall net coefficients for the roof suction (or assume a triangular load distribution) or do you go through each of the individual roof zone suction contributions to OTM?

 

[Eng16080]

There's a discussion on whether or not to include the roof suction loads (pink in sketch) in the total overturning moment.

I don't understand what their argument would be for neglecting it. It clearly adds to the clockwise overturning moment about point B.

 

[OffshoreWindStructures]

Maybe too many years spent working on unclad structures (pipe racks), etc, to consider roof cladding suction actions.

 

[m7rhman]

All the loads should be considered simultaneously for stability check. If it enclosed, with flat roof, will develop negative pressure. ASCE 7 gives leeward coefficients as a function of length to width, and roof pressure as a function of height to length (the length is parallel to the wind direction under consideration). See screenshots for the coefficients.

Notice, the maximum value coincide with what you mentioned (0.8 + 0.5).

For roof

Notice also, there is the simplified method in chapter 28. And by using this you treat the container as a building (chapter 27 and 28). One could debate you should use chapter 29 (appurtenances), which has validity. I believe you should give it a look and use your judgement.

 

[JStephen]

"I don't understand what their argument would be for neglecting it. It clearly adds to the clockwise overturning moment about point B."

I don't see anything in the original post that indicates where those loads are coming from.
But if it's subject to a building code, that code will generally specify what (and where) those loads are.
Where I've run into this, is with vertical cylindrical tanks. ASCE 7-10 and older figure wind loading based on projected area only. ASCE 7-16 and newer figure wind load on projected area and also uplift on the roof. The result is that there is a pretty sizable difference in wind overturning between the two standards. Or more to the point, if the code being used doesn't specify loads on the roof, then presumably there are no loads on the roof, per that code.
It's hard to tell from the diagram, but there are potentially wind forces on the bottom of the container as well.

 

[OffshoreWindStructures]

JStephen, it was more of a loads from first principles discussion than a loads from code discussion as it was a general topic rather than particular design to a particular code, and for this example you can assume there is no wind flow (so no suction forces) on the bottom of the container.

I'm not familiar with ASCE requirements and version changes but i did find this guidance doc from ASCE press for ASCE-7-10 that only considers windward and leeward pressures for overturning moment of buildings, which is consistent with JStephen saying that any contributions to OTM by roof suction was ignored in earlier versions.

edit :

I also found the same doc for ASCE-7-16 and the approach that example does not include roof suction in overturining for buildings even if it does for vertical tanks as JStephen says.

 

[GeorgeTheCivilEngineer]

I don't think the argument is to fully ignore it, the question is whether it is 'worthwhile'. Given the other half of uplight mostly counteracts the uplift overturning and the roof length is small, the overall effect is likely to be negligible and can be ignored.

I'd do a quick order of magnitude check on what the actual uplift overturning is, and if it is, as I suspect, negligible, then I would ignore it but make a note that is the decision I have made.

Do you use overall net coefficients for the roof suction (or assume a triangular load distribution) or do you go through each of the individual roof zone suction contributions to OTM?

From memory there's a minimum length to create zones so I think it's just a single zone. Don't quote me on that though...

 

[OffshoreWindStructures]

I'm not sure what you mean when you say the other half of uplift mostly counteracts the uplift OTM. It's my understanding there should be at least one case where all roof uplift acts in the same direction so there shouldnt be counteracting uplift.

For clarity, here is some examples from online calculators that match with my understanding of wind suctions and zones for a 12m x 2.5m x 2.5 m shipping container. The final values differ due to different pressure/suction coefficients chosen.

and more detailed version of another online calculator for walls and roof


[URL unfurl="true"]https://res.cloudinary.com/engineering-com/image/upload/v1729850035/tips/Calculation_of_wind_pressure_loads_-_Eurocode_1-_ShippingContainerRoof_ichoem.pdf[/url]

[URL unfurl="true"]https://res.cloudinary.com/engineering-com/image/upload/v1729851645/tips/Calculation_of_wind_pressure_loads_-_Eurocode_1-_ShippingContainerWalls_peqgxr.pdf[/url]

 

[RPMG]

For a simplified version, use wall pressure only with Cp = 1.3 (0.8 windward + 0.5 leeward). Assume the container dead load negates the roof uplift.
For greater accuracy, include roof uplift with Cp = 1.3 and minimum dead load combinations.
The overturning case will always be with L/B < 1.0 and h/L = 1.0 approximately for a container, so I wouldn't attempt a more exact solution.