Strength of Existing Metal Roof Deck

[KYStructuralEIT]

As part of a project to support a 4,000+ lb mechanical unit on an existing building, I will need to check the metal roof deck.

The unit sits on 1 1/2" curbs that will be attached to the existing wide-rib 1 1/2" x 20GA metal deck. In turn, the curb load will be supported by new beams.

My question is: how can I check that the metal deck won't crush under the curb loading running perpendicular to the flutes (275 lb/ft in this case)?

I am a recent graduate looking to gain experience from this group.

 

[JedClampett]

This is almost an impossible analytical problem. But as a newbie, you should always try to think of issues like these in practical terms. Four thousand pounds is about the weight of a Buick. Would you want to drive a Buick on those flutes? The load distribution is not through wheels for a mechanical unit, but a skid or frame, but the size of the unit is probably smaller than a car.
Provide additional members underneath the frame and a curb to transmit the loads to those members. Either the manufacturer of the unit or a curb company can design a curb to carry the weight of the equipment.

 

[spats]

Agree with JedClampett... do not put that size of a unit on the deck.

 

[MiketheEngineer]

Agree - the deck would probably deform. There are a few ways around this. Build a curb that wraps around the hole you put in the deck. Use angles that are taller/wider than the deck or put stand-offs through the deck and attach new framing to them

I have even seen new plates/angles fabricated that actually matched the deck profile and installed. Seems like a pain to me!

 

[KYStructuralEIT]

Thanks for the responses.

The new mechanical unit will be supported by two new W16 beams, installed below the metal roof deck.

The mechanical unit will sit on top of a curb that will transfer the load to the new beams.

The issue is the connection of the curb to the metal deck and if the deck can transfer the load effectively to the beams below without crushing. Or will I need to specify openings in the deck to attach the curb directly to the beams?

I appreciate the advice.

 

[KYStructuralEIT]

Thanks MiketheEngineer.

It seems like the overall opinion is that the deck is not strong enough to transfer the load effectively. I will design stand-offs to penetrate the deck and attach to the beams.

 

[firai]

Your deck catalog should offer web crippling strengths for bearing by a wall or a curb. However, I would forgo that altogether and design a penetration through the deck like others have suggested when you have equipment that's 4k+.

Structural Design Engineer
New York, NY

 

[hokie66]

Rather than cutting through the deck, you could install packers between the beam and the deck pans.

 

[flash3780]

Can you gin up a quick 2D plane-strain FEA? Wouldn't have to be a terribly complicated model - just the profile of a single corrugation. Finding the loading is pretty straightforward.

 

[firai]

hokie66: I've never seen that done before, though it certainly sounds feasible. Do you know if it's hard to pack the underside of flutes?

flash3780: AISI actually has a direct analysis method that gets better strength predictions than using FEA alone. I don't think it's really worth the time to do the analysis though for one piece of equipment.

Structural Design Engineer
New York, NY

 

[hokie66]

firai,

Not difficult at all if you can get at it. Access to the underside of the deck is often restricted.

 

[a2mfk]

I've never designed anything to bear on a roof deck other than people walking, or maybe some lightweight ballast system.

Guys got you covered on this, bypass the deck to get to the WF beams. You also have to design for overturning and uplift due to wind/seismic, and if you are in snow country this will cause drifting around the unit.

 

[Hokie93]

The crushing capacity of steel deck is tabulated in Deck Design Data Sheets #36 and #37, formerly published by United Steel Deck, Inc. Now that USD is Canam, I don't know if the Data Sheets are still produced or otherwise available.

 

[a2mfk]

Dang, another Hokie on this list? Can't you turkeys spread your wings a little and come up with some other names?

 

[hokie66]

We just like to irritate Seminoles. Serves you right for wearing turkey feathers in your headdress.

 

[SteelPE]

When we install rtu's and have access to the decking we will use a channel with the width of flange equal to the height of the deck. We place the toes of the channel down and from support to support. This way you don't have to worry about crushing the deck. Of course it won't work unless you have access to the deck.

 

[a2mfk]

Thats a decent detail too SteelPE...

Hokie66- if you graduated in '66, then all these other hokies owe you an Aussie nickel every time they post. Official ACC message board rules.

 

[hokie66]

Good one, SteelPE. Now that you have reminded me, I have done it that way.

Sounds fair to me, a2mfk.

 

[Ron]

OP mentions a 1-1/2" curb. I'm assuming he is talking about the framing angle around the opening, which will have a wooden curb attached to it...if not, 1-1/2" in not enough for the roofing considerations.

Loading perpendicular to the flutes is not as much of an issue as loading parallel to the flutes.

Support from bottom as others have noted. If you are using a pre-fab curb section, most will span the deck flutes without issue and the load distribution is such that the deck corrugations are not deformed much, if at all.

As hokie66 mentioned, flute packing can be done just make sure you don't create a condensate dam in the process.

 

[Bobber1]

A little off topic but you might want to make sure the mechanical engineer or whoever specs this unit consider placing it on vibration dampers. These will be above the unit base frame so it shouldn't affect your mounting to the deck.

Sometimes a roof top unit can make a metal deck vibrate like a drum and sound accordingly.

Bob G.