What is wall tie in warehouse for

[JackEngTip]

Hello everyone,

I am relatively new to warehouse design and hope someone can help me regarding this.

Above is the scheme of structural plan.
My questions are:

1. What is wall tie (normally PFC in 90 degree axes angles) for? My understanding is to work as a tie in 'strut & tie wind truss' to resist wind load. But why don't we put wall tie in orthogonal grid? instead we put racking angles (normally EA) with bolts to precast concrete panels? I undertstand racking angles can work to carry purlins but dont we need tie to resist wind in orthogonal direction as well? How can EA with botls to panel work as tie? If this is possible why dont we simply use racking angles all around?

2. How does wind load get distributed? I would say in this case wind goes to base and roof (the panel work as a one-way slab that runs vertically). But cant the wind load be transfer horizontally and then goes to other panels at ends, which have great in-plane capacity so we can get rid of wall ties and the like?

3. How about a structure like this?

portal columns and glasses between columns. Do I apply the wind load as line load on columns or roof beams and why? If wind applies on roof beam does that mean noramlly UB should be avoid for roof beams due to the laterally deflection in beams? What kind of section do you normally use? SHS?


Thank you very much.

 

[JAE]

Bump - anyone?

Check out Eng-Tips Forum's Policies here:
faq731-376

 

[Ron247]

Bueller? Bueller? Bueller?
I do not understand some of the terminology and am not following what you need help on.

 

[JackEngTip]

Thank you JAE and Ron247.
They call it wall tie in Australia. No idea if we are talking about the same thing. It is simply just members (normally PFC) running along precast concrete panels (abut) and connects with RA (EA Sections normally) and roof beams. So despite the difference in terminologies, can you see how this structure is gonna work? Or how warehouses with precast panels are designed in your countries?
Thank you.

 

[civeng80]

Your question is a bit confusing. Are you designing load bearing concrete panels or a portal frame with non loading panels ?

 

[human909]

The overall design isn't clear. But it might be because the walls on the left and right are considered to be constrained by the purlins.

If the purlins are designed with sufficient capacity then it could work. But you don't want to get that wrong.

I've seen a portal frame building have its entire end wall blown in because some ignorant engineer didn't have any compression path from the from portal face apart from the roof purlins. The purlins bent, the facing column moved and the foundation cracked. The thing that saved the the entire wall was the fact that the large roller door came off its tracks and blew in.

 

[JackEngTip]

Hi civeng80 & human909. Sorry for the confusion. The second picture is for the third question only (I simply wanna know how wind load is distributed in that case), nothing to do with the warehouse. There is no portal frames in warehouse, just panels and roof structure.

 

[human909]

Hi civeng80 & human909. Sorry for the confusion. The second picture is for the third question only (I simply wanna know how wind load is distributed in that case), nothing to do with the warehouse. There is no portal frames in warehouse, just panels and roof structure.

Depends like all structural analysis it depends completely on how the panels are supported and fixed. You are going to have to flesh out the fixings more to get an idea.

 

[civeng80]

Sometimes in tilt up wall construction the roof beams are not at the joints between the panels, so that a steel channel is used to tie the panels together. If the roof beams are at panel locations then some engineers dont bother having the channel ties. It is debatable whether thay are needed in this situation, but I still put them in, in case the builder wishes to change the panel width and also they complete the steel bracing system for the structure.

I hope that answered at least part of your question.

 

[Agent666]

The channels are there to take the face loading on the precast panels back to the roof bracing nodes. Perhaps due to the smaller span for the EA in the end wall, the designers try justify the EA spanning by itself between the eave and the ridge, but depending on the spans I doubt it would work too well (EA not good in flexure), and you would require a channel in the end walls as well. I wouldn't rely on the cladding to act as a diaphragm because well its not a structural member with 50 year design life, though arguably it does something in practical terms.

The channels and angles are also used to distribute the roof bracing loads along their length to the precast panels via distributed fixings into the panels usually.

 

[JackEngTip]

It is just some precast panels with cast-in plate to connect themselves together and dowel bars to footings.

 

[JackEngTip]

So if I understand you correctly, if roof beams are at the joints of panels or if there is only one panel, these wall ties are not needed because we can use the panels themselves (with reinforcement) as ties?
What about the orthogonal grid, how can we use EA considering it has very weak capacity in bending? Is it like what human909 said, purlins are used as the support for the EA, which means the span of the EA is quite small so bending shouldn't be a problem?

 

[JackEngTip]

Hi Agent666, seems like it is quite a normal practise here to use channel on one direction and angle on the other. I haven't seen channel all around to replace EA (EA will be used to carry purlins too). I am not quite sure the span of EA you refer to here. Let's say each EA has bolts to panels @1.5 meters. Is 1.5 meters the span? (I don't think so as EA provide the lateral support for panels not the other way around). Or is the purlin spacing (lets say 1.2 meters) considered as the span of the EA when carrying out the lateral bending calculations? (this is the only way I can see how EA is sufficient here because otherwise if we take the entire span of EA from start node to end node of the roof truss that will be too long for EA and its lateral bending will fail for sure)

 

[civeng80]

I use the channels even if the roof beams are at the joints so that the whole roof bracing system is steel, but I have seen some engineers do away with them completely.

 

[JackEngTip]

I can understand how channel works but still am not sure how EA works in another direction.

 

[civeng80]

In a tilt up building with many bays say 4 or more, the purlins between the struts carry the load to the roof beams in the weak axis direction (assuming reserve capacity in the purlins to act as struts) and this would normally be enough to transmit the forces to the side walls. The EA then span between the purlins.
If you have just 1 bay then the EA may have to be relaced with channels and the bending span would be the span between the struts. In this case 1/2 load can be used on the end walls where again the purlins would need some extra capacity as struts.

I hope that answers your question.

 

[JackEngTip]

Thank you very much Civeng80. you are a legend. I think I know how it works now (although I haven't seen anyone running a buckling analysis on purlin for lateral load, maybe the load is too small so it can be ignored.) I have also seen people using a plate on top of the panels joints to get rid of channel and I think their idea is to use the panel itself to resist the lateral load.
My last question is irrelevant to the warehouse design. It is just a general question regarding the wind load distribution. The above case we apply wind load on roof beam members (either channel or EA). If I have say, a rectangular room that has 4 external glass wall with light weight roof, and I am using portal frame to resist wind load in both directions. My question is do I apply the wind on the column or the roof beam and why? ie. Does wind goes vertically or horizontally? Or do I need to do FEA and use area load to determine as it is stiffness based? Thank you.