warehouse design question

[n3jc]

Greetings
I'd like to hear some opinions on this structure (tire warehouse) that has:
-timber roof (rafters supported on timber beams),
- timber beams are supported on RC frames,
- there is a RC slab above ground storey above entire structure – except the first field which is empty,
- walls (concrete or masonry) are 300 mm thick with only few openings.

I am wondering what is the best way to design this one regarding horizontal - seismic forces.
Since RC slab is not above whole structure (first field is empty) this causes weakness in transverse direction. RC frames will support timber roof and will be designed as moment frames so they can also take transverse forces but I was thinking about bracing in the roof plane, which will transfer transverse horizotnal forces above RC slab to both external transverse walls. Horizontal forces in other/longitudinal direction should not be a problem since I have two long external walls with only few openings so bracing in the roof plane will transfer this forces to longitud. walls. Timber beams which support rafters are part of bracing (compression forces) so maybe steel is a better option than wood.

I am wondering what is the best way to design this – especially for horizontal forces in transverse direction? Any suggestion?
I know I can design this structure as RC frames with fillers but I think there is no need for that. I can also use steel frames instead RC frames above the slab.

This structure should not be a precast structure!

Id like to get a design that is as simple/cheap as possible. What do you suggest?

 

[Ganesh Persaud]

I would suggest doing a steel portal frame, either an IMF or SMF frames. For the slab, 4.7m you don’t need any filler beams however during construction, I would recommend shoring until the 28day strength is reached. In my sketch, I have the blue representing the main frame members and the red representing the wind bracing. Then you could have z-purlins along the length of the building to support roof sheeting. Remove the timber framing.

 

[KootK]

I see merit in all of the ideas proposed so far but I feel that, with that rigid slab connected to a transverse shear wall on only one end like this, all of the options presented so far will have a serious concentration of seismic force at the columns circled in red below. In the sketch below, I'm suggesting that one simply own that behavior, design for it, and strip away anything else that may be extraneous in the interest of cost savings.

While I wouldn't recommend sole reliance on it, you're going to have wickedly stiff three sided shear wall system surrounding your elevated slab. Odds are you could do nothing but attach the slab to the walls for shear transfer and you wouldn't have any problems.

 

[dold]

A concrete frame might be the last option I'd go with for "...as simple/cheap as possible." Why can't you use masonry walls and just a normal wood roof diaphragm?

The floor diaphragm still needs attention. Concrete slab is ideal for a warehouse I would assume. Plenty of stiffness there for a 3 sided rigid diaphragm analysis.

Are there 3 basement walls or 4? If 4, I'd run a rigid diaphragm check on the floor slab and see what it does with the loads. If 3, design the column for bending or do what KootK is talking about.

 

[n3jc]

Thank you both for reply.

Ganesh Persaud I was thinking about something like that but they are not that keen on the steel.

KootK I agree with what you wrote. Just one thing - if I brace all fields (not just the first one where slab is missing) I transfer all transverse horizontal forces (above RC slab) to both outer walls . But if that is not a problem, bracing all fields (excpet 1st one) is unnecessary since I have RC slab right? Im kinda worried about forces in bracing (steel diagonals) and anchoring this forces. My columns can also be just 300 mm wide so that might be an issue too. I ll make a model and see...

thanks

 

[KootK]

In terms of structural performance alone, I like both your idea and Ganesh's idea better than my own. More redundancy etc. I was basically exploring how little structure one could provide and remain viable. Still, even trussing the whole roof, I wouldn't expect the bracing forces in the first bay to be too much lower than they'd be with my proposal.

 

[r13]

My thought, the roof braces can be eliminated if the RC frames are longitudinally braced (by walls, beams) with minimal relative movement between frames.

 

[n3jc]

This is what I came up with. As far as seismic goes this gave me the best results, the question is how expensive this is. I like RC roof slab since it really connects structure together and there is no detail issues with steel bracing etc.

One more question regarding cost: compering RC moment frames that supports roof compared to the steel frames? RC is much cheaper material, but a lot of work with reinforcment and formworks... steel and steel connections are expensive but is quite fast to install.