Any tips / sources for warehouse design / bid?

[AaronMcD]

We are asked to bid 140k sqft of warehouse space, 2 buildings, 2 stories each. Normally do residential and hospitality. Any tips on how to bid, and online sources or personal tips for design? (seismic area)

Anyone have special design considerations that might not be intuitive? Mainly things like how to deal with the huge diaphragm (e.g. do you break it into subdiaphragms? Is there a length limit between lines of seismic resistance?) or foundation (foundation ties? cross footings or in-slab ties? Loads to design ties for?).

Our usual projects do not have much optimization. Where would you consider going over minimum sizes? Any typical construction methods and/or details?

Any typical construction methods that may not be obvious? All steel, vs concrete or tilt up first floor? Hot roll vs OWSJ at roof? (interior columns, 30 ft spans)

Kind of a broad open-ended question, I know. I can probably find all the info on design eventually, but any head start is appreciated. First step is bid.

http://www.hellodwell.design

 

[dik]

There's lots of stuff to consider, main ones likely wind and/or seismic (if in a seismic area) and the lateral resistance of same. What foundations are common to the locale? Slab on grade for main floor? I'd probably use simple steel framing for the second floor and plastic steel design for the roof. Depending on your snow loads, in these environs 1-1/2" 22 ga deck is most cost effective and is good for a little over 6.5'. OWSJs are most cost effective and a span to depth of about 16 or 18 are most cost effective. Columns could be HSS, but connections for the second floor may be less costly with W sections. Roof framing should use continuous beam construction. Is there a fire resistance required? Bay spacing could be something like 40' x 25' or 30'. What is the use? Is there heavy loading on the second floor... may require smaller bay spacing. Can you use OWSJ for floor? or steel beams and possibly composite construction? Any forklift traffic on second floor?

You should look at the framing system to start with, including foundations; this will likely determine the economics. The 'tightness' of the design doesn't really come into it. What is the cladding? and roofing? What is the coating system used? A properly selected coating system can save money. Is it sprinklered? Any equipment on the roof? Just too many questions...

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[jayrod12]

I agree with dik's points for the most part.

The one where I have caution so I always have a conversation with the owner before I get too into the design of the roof, "What are the chances you're going to have tenant change overs requiring new/different mechanical units on the roof?"

I ask that because when you have a gerber style roof system, or continuous beams, sure you save money on the steel weight for the specific loading you've designed around originally. But once the loading changes, there tend to be cascading style reinforcing required even in bays where you aren't necessarily changing the loading. So often I'll just keep the simple span beam intent.

I've worked on too many projects recently where the superstructure has gone out for tender before the roof equipment is finalized. I've even had projects recently where they have changed the roof unit layouts and weights after the roof is constructed. At least with simple span beams the analysis and detailing of the reinforcing is simpler.

 

[AaronMcD]

All good points, and even more questions.
This is California, no snow. Schematic bay sizes are 30x27.5, and not centered at the ridge. I'd think these could be increased a bit and maybe centered on the ridge.
Probably could be forklifts, I imagine that's the easy way to move stuff around in such a large space. We were told typical warehouse loads, and indoor farming.
Metal siding and roof currently shown on schematic design.
Slab on grade probably. Not sure about foundations, haven't seen the geotech report yet. On farmland about 1/2 mile from the river. Most of our residential projects are typical inverted T footings, and the occasional mat slab on piles.
What does "continuous beams" mean? Is that just where each one cantilevers to pick up the next, or moment connections?

http://www.hellodwell.design

 

[MotorCity]

Check client's insurance provider requirements.

 

[canwesteng]

The roof girders will be continuous beams over the interior columns - the joists are supported on the girders. Gerber girders is also an economical system.

 

[dik]

In 50 years, I don't recall ever doing a Gerber framed roof.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[AaronMcD]

I never heard it called that, but it looks like it's just cantilevering the girders. You've never cantilevered girders to pick up the next beam near the inflection point? Pretty sure I've done this on single family homes. Or is it only "Gerber" if it's systematic with smaller drop in beams?

http://www.hellodwell.design

 

[dik]

I think the latter...

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[JLNJ]

We just did a 400,000 sf Gerber style roof. I think you get a significant steel savings, but bracing the negative moment regions over the columns is a real pain in the neck, especially when you have bar joists.

And deciding on the patterned loading used to determine the worst case for the cantilever condition requires some judgment.

 

[dik]

With plastic design, you likely would have saved more, specially for roofs with uniform column spacing. Pattern loading isn't so much of an issue with plastic design, it only affects the connection moments.

Both the building roofs were plastic design, the larger one was about 400,000 sq.ft.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[AaronMcD]

I don't think patterned loading would be very difficult, especially with no snow load and a repetitive roof.

What is "plastic design"?

Also, what is "typical" for direction of steel joists on a sloped (~2:12) roof? It seems joists parallel with ridge would be the norm so the metal deck slopes to drain, but how to connect sloped deck to joists?

http://www.hellodwell.design

 

[dik]

Plastic design is where the failure mechanism determines the limit load. With some framing systems, for example continuous steel beams, the limit load goes beyond the elastic capacity of the beam.

https://onlinelibrary.wiley.com/doi/10.1002/9783433606483.ch5

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[phamENG]

Metal deck is not a roofing system, just part of the structure that supports it. So you run your joists perpendicular to the ridge with sloped seats, put the deck on, then your roofing system (often some amount of rigid insulation, a roof protection board, and then a membrane).

Plastic design is a method of analysis that looks at the failure mechanism that causes collapse. If you consider a beam with moment connections: in the sort of design you're accustomed to using, you would calculated the maximum moment (moment at the fixed end) and design for that. But if you form a plastic hinge (full yielding) at the fixed ends, the beam hasn't collapsed. The ends can continue to carry that amount of moment as the load on the beam increases, you just continue to get rotation and the moment at mid-span increases until it yields. Once you're formed a failure mechanism (the remaining structure is no longer stable), that's it. It lets you get a lot more out of the structure using less material. It's particularly well suited to warehouse roof structures that aren't usually deflection sensitive.

My steel design professor was a big proponent of it, but I've encountered it very few times "in the wild." dik is an exception to that rule.

 

[dik]

...been using it for 50 years, and haven't looked back. I did my first big warehouse about 50 years back for a prof using elastic design. He asked why I didn't use plastic design, and since then I was 'captured'. When I first started, I was told that plastic design was about 20% more expensive... not true, today, or back then. Less weight and fewer pieces to handle. Another myth is that you had to be careful about the placing of the members so the cantilevers worked for erection... not generally true.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[AaronMcD]

Ah, yeah I did that in college but never used it in real life. Are you not worried about permanent deformation?

Sloped joists makes sense. We would obviously need a ridge beam in that case. I haven't seen many warehouses up close, I guess I thought most just use metal deck as the roof material like a big shed, and put any insulation inside.

http://www.hellodwell.design

 

[dik]

Catch...

[URL unfurl="true"]https://res.cloudinary.com/engineering-com/image/upload/v1641846020/tips/205_52_abfhmv.pdf[/url]

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[phamENG]

PEMB's do it like that. Rigid frames that run across the building, then purlins run between the rigid frames parallel tot he ridge, and the roofing is screwed straight to that. But then they have in plane rod bracing since they don't have a real diaphragm (usually). But if you're using OWSJ it's best to have a beam down the ridge and slope your seats. Pay close attention to that detailing. The joist suppliers have limits on various seat types, sizes, slopes, etc. You'll also need to be careful with the seismic detailing to transfer your shear over the ridge since you'll have a discontinuity in the steel deck diaphragm.

In theory, you'd never need that full plastic capacity. You use it with factored loads, so you're looking at ultimate loads - the loads you never expect the building to actually see but those for which we design it just in case. So under normal service loads it will probably remain elastic anyway.

 

[phamENG]

dik - nice! My prof knew Beedle, and I have no doubt that paper influenced him.

 

[dik]

Nope... generally elastic for service loading and if any yielding occurs. the structure behaves elastically up to the loading that caused the yielding.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik