Wind Code Combination Factor

[asixth]

Hi guys,

I'm doing a roof bracing design currently and some of my bracing struts are working a bit hard.

Does anyone use the combination factor (Kc) as set out in AS1170.2 Cl.5.4.3?

In my situation, the windward and leeward walls are contributing to the element force, I haven't considered friction drag across the roof and the horizontal component of normal roof pressure is negligible.

If I use Kc as set out in the code, I can actually reduced my member force by 20% (Kc=0.8).

 

[hokie66]

It is legitimate to use the combination factor, provided your ratio of windward to leeward pressure is less than .75. I would tend to use it for the horizontal bracing trusses and wall bracing, but not for any distribution struts across the building, as compression forces in these struts are already reduced by leeward wind suction.

Having said that, for many buildings, an area reduction factor of .8 can be used for the roof bracing, and in such cases, the combination factor is 1.0.

 

[apsix]

Hokie
Can you tell me where the 'ratio of windward to leeward pressure is less than .75' rule comes from?
Thanks

 

[hokie66]

apsix,

I was just paraphrasing the condition in Table 5.5. It says "where wind action from any single surface contributes 75 percent or more to an action effect" that the combination factor is to be taken as 1.0.

 

[sdz]

As hokie66 says check that Ka * Kc >=0.8
It's easy to double count.

Depending where the structure is you might be able to use the wind direction factor Md=0.95 as well.

 

[asixth]

Thanks for all the replies, how come no-one responded to my question regarding the Earthquake code?

hookie66,

I am taking my windward and leeward co-efficients as +0.7 and -0.5 respectivley, so I guess the windward wall will contribute to 0.58 of the member force. I don't understand what you mean when you say "as compression forces in these struts are already reduced by leeward wind suction.". I am effectivly modelling a large 2-D truss acted on by a horizontal windward and leeward force that are both acting in the same direction and therefore, both contributing to the compressive force (referring to the top chord of the truss).

I don't think I can use the area reduction factor because I am not modelling any contribution of lateral force from the roof. Saying that, I didn't think you can use a Ka for windward and leeward walls.

However, I do understand that I cannot use a Ka on top of a Kc and have a total reduction less than 0.8.

sdz,

I have always used an Md of 0.95 when considering the lateral system of the building and only use a full Md (e.g. Md=1) when I am designing individual component, like purlins and blockwall cladding.

 

[hokie66]

asixth,

I stand corrected about the area reduction factor. You are correct, it doesn't apply to windward and leeward walls. I don't really know why, but it could be because the code writers just wanted the combination factor to be the controlling reduction for racking.

As to my other comment, what I was talking about was struts across the building, not the truss members. I tend to call the truss members chords and webs, and compression members running parallel to the ridge are struts. In big buildings, it is common to use a truss on each end and distribute the force between the trusses by struts (or sometimes by the purlins). Sorry if I made your question more complicated than it was.

 

[apsix]

asixth

"I have always used an Md of 0.95 when considering the lateral system of the building "; By Cl.3.3.2 that's only applicable for regions B, C & D (which counts me out).

I read but didn't respond to your earthquake question as I have no experience with the current code.

 

[OzEng80]

asixth

I would also use Kc=0.8 and Md=0.95 for your arrangement. I believe this is generally appropriate for determining raking forces on most buildings.

One question though; isn’t it possible for a bracing condition where all the lateral force could be applied to 1 wall? Ie dominant opening on side, lee or roof resulting in internal suction on windward wall. This scenario could also result in +0.7, -0.5 on one surface and Kc would be required to be 1.0 and a worst loading condition for the ‘struts’ regardless . This is a very likely scenario in regions C and D for buildings without cyclone rated, glazing, doors and cladding.

 

[asixth]

OzEng80,

The design wind event is 60m/s which is essentially the gust speed you would expect in a cyclone and I guess it is a very real scenario that the roof cladding will be lost during this event.

Saying that, I have never considered internal pressures where I am designing global elements. If an internal pressure is acting on one wall that is increasing the design action in an element, than that the same internal pressure will also be acting on another wall somewhere else where it is decreasing the design action.

apsix,

My apologies, I forget sometimes that the world does not revolve around Brisbane.

All,

Thanks for your responses. I have justified the use of the Kc factor to myself by considering the turbulent nature of wind which creates the unlikliness of full external pressure to be entirely developed on both the windward and leeward walls simultaneously (without getting to in-depth with fluid mechanics).

Just a final question:

Is the Kc factor applied to all surfaces that contribute to the force in the element under consideration. Take a rafter, it forms part of the lateral system so it sees a significant compression forces, along with bending moment that results from uplift over the roof. Is the Kc applied to windward and leeward and roof in this scenario?

 

[OzEng80]

Asixth

I think your lateral stability statement is common practice, but it would be nice to have a literal reference somewhere. The internal suction only shifts the -0.5 from the lee wall to the windward wall it doesn’t cancel it out (ie +0.7 windward -0.5 / -0.5 lee -0.5). This is a nasty arrangement for the bracing system (large strut forces) and disputable kc value.

Do you take Cp’s of +0.7 & -0.5 for a single wall design?

The Kc factor is applicable to all elements.

Cheers

 

[sdz]

You must still consider internal pressure when designing global elements. e.g. it will have a dramatic effect on a portal frame. Analyse a portal frame with internal pressure as a separate load case and try it with Cpi= +0.7, -0.65 and you will see what I mean.

 

[asixth]

OzEng80 + sdz,

Honestly, I always went about my design ignoring internal pressure because much like a balloon sitting on a table, internal pressure creates a zero net force.

However, the positive internal pressure inside the building will cause elements in the roof frame to go into compression (assuming Cpi=-ve). This is an something that I have not considered in by design.

Thanks guys.

 

[sdz]

Internal pressure may not affect the overall horizontal wind load but it greatly affects the load on individual elements and may change the distribution of base shear between supports. You should be considering the possible range of internal pressure depending on where dominant openings could be, especially in cyclones where wind direction can change 90-180 degrees during a single storm.

What type of structure are you designing anyway? Braced frame, portal frame, shed, house?

 

[asixth]

Braced roof with precast walls.

 

[OzEng80]

Given that it appears that the +0.7 cpe, -0.5/-0.65 cpi is an appropriate lateral stability case, is the Kc factor applicable (all load on one surface)?

I have read the appendixes of a couple of building design type handbooks/standards (eg wind loads for houses) and a Kc of 0.8 is adopted for design.