Metal Building with Basement

[stiffener]

Our client is requesting using a single slope metal building in order to save money on his next project. The building is also to have a full basement with a 10' floor to floor elevation. The floor framing is to be a concrete on metal deck supported by open web steel joists. The issue that arises is the control of the horizontal thrust from the building. Typically with a slab on grade application we would call out a strip footing or grade beam between the columns to resist the forces.

In this application we are thinking of sending the thrust loadings through the deck. The joists are to be connected to the foundation wall with an embedded bearing plate. My question is, what steps need to be taken in order to distribute the horizontal thrust to the joists between the columns. If this can't be done efficiently, we are thinking about having a single steel beam at each column pier location that will span the width of the building.

 

[jopalu]

Could you run a tie rod through the concrete on metal deck? I doubt the steel joist manufacturer will allow tension, applied in this manner, to the joists.

Another thought ... Can hairpins along with WWM in a 3"+ concrete slab on metal deck accomplish the same as the SOG procedure? (This is a question before a suggestion...)

 

[SperlingPE]

Design your concrete piers to take the horizontal load. This may or may not be practical due to size of load and space constraints for the pier. Hairpin into the slab for relatively small loads. Specify a long enough rod and be sure slab reinforcement is placed properly to get the desired results.

 

[stiffener]

Thanks for the responses. What about having a steel beam at each column location that would be tied into the pier and span the width of the building? I have four bays along the width, so the force would have to pass through a column connection.

 

[WillisV]

Joist manufacturers are certainly capable of designing their joist for an applied axial load as long as this load is clearly spelled out on the design drawings.

 

[HouseBoy]

What are the approximate dimensions of the building?
Are there interior columns at the upper floor? (just wondering as it would affact the magnitude of the thrust)
Could beam lines (at the floor) be arranged to coincide with the frames? (doesn't sound like it but again, just wondering.
I don't see a problem having joists designed fo axial load. The connection at the joist bearing would likely be something special but I agree with WillisV.
I'm not sure I understand your comment about four bays and having to pass thru a column. Maybe the building dims would help clarify.

 

[stiffener]

The building is 80' wide with a length of 180' with metal building columns at approximately 20' on center. There are no interior columns to the roof in order to minimize thrust(I wish there were).

The first floor framing consists of four 20' bays along the width with steel joists supported by the foundation wall and three steel beam support lines. I do have the columns for the first floor aligned with the metal building columns. That is why I suggested having tie beams at each column that would tie into the piers. The tie beams would have to connect to the interior columns and pass the tension force through the column to the next tie beam. For the connection of the beam to the pier, I am thinking of plates that would connect the bottom flange and the web to the pier. Any ideas? Thanks.

 

[jopalu]

Stiffener;

A steel beam is definitely capable of serving as, what I will refer to as, a tie rod. (or tie beam)

Many things to consider:

-What would be the span of the beam?
-Is deflection a concern?
-You have four bays along the width of the building.
-I don't see a problem designing the thrust to pass through one or multiple columns, however this would have to be approved by the metal building designer - either by their design or your design, approved by the MB designer. Curious; At what elevation will the interior columns bear? 100'? 99'-4? 90'-0?
-Will your tie beam be, in any way, part of the concrete floor?

There are many, many ways of designing a satisfactory tie-rod system ... sounds like details may be key ...

 

[stiffener]

The tie beams would span 20' from column line to column line. I was thinking that I would have the beams at the same elevation as the support beams, therfore 2 1/2" below the concrete deck and not tied into the deck or the slab. Deflection would not be a factor then. I would have a tie beam line and an interior column at each MB column.

The MB columns bear on the foundation wall at elevation 100'-0" which is the same elevation of the first floor. The interior column bear at elevation 90'-0".

I am most concerned about the connection of the tie beam to the pier. The connection would have to tie around the anchor bolts to transfer the loading.

 

[jopalu]

stiffener;

The magnitude of your thrust will govern the size and required length of anchor-bolts.

Beyond that:

Could threaded tie-rods be thread-and-nut anchored to a plate/bracket, designed by you, behind the anchor bolts, project out of the front of the concrete pier and then, thread-and-nut fastened, to a steel bracket/gusset, designed by you, shop welded to each side of the tie-beam web?

Do not neglect considering what effects this low thrust placement may have on the anchor-bolts and concrete pier.

 

[DaveAtkins]

I suggest a plate on the vertical inside face of the pier, with hook bars extending almost to the outside face of the pier, welded to the plate. A single plate shear connection should be able to transfer the thrust from the pier plate into the tie beam (check the bending of the pier plate between the hook bars).

All of this depends on the magnitude of the thrust force, of course.

DaveAtkins

 

[jopalu]

stiffener,

Please forgive me for pushing a suggestion that seems to have been discarded early in this thread. I only bring this up because it seems that you would be going through a great, unnecessary expense with these tie beams. If you do proceed with ties beams, I suggest DaveAtkins solution over my previous suggestion. His pier to beam connx seems far simpler.

May I ask what your thrust load is? Since you have interior column(s), the thrust should not be exceedingly great as opposed to a clear-span low-profile building.

Another question is what concrete thickness over top of metal deck will you have? Are you sure a simple, round tie rod cannot be accomplished through the concrete over metal deck? I would suggest 3" concrete over top of metal deck. Would an additional 1-2" of concrete be that much more of an expense than the apparent 80' length of beams and details of force-transfer-through-columns, at each rigid frame line?

Estimate the cost. I could be wrong ...

 

[stiffener]

jopalu,

I don't have the actual thrust numbers yet. I have a preliminary maximum thrust from the MB of 46k. The main building does not have any interior columns up to the roof, only the first floor framing has interior columns. The tie beams would be in 20' segments between the interior columns.

I plan on a 4" concrete slab with appox. 3" above the deck. Would the tie rod present cracking issues with the concrete deck when loaded?

 

[jopalu]

Stiffener,

Please forgive me! I thought the interior columns were MB. Maybe your tie-beams are best ...

 

[archeng59]

Stiffener, if you have a basement, is it possible to have counterforts at each rigid frame to resist both the lateral frame reactions and the soil loads?

 

[stiffener]

That is an option.

But isn't that essentially the same method as sizing a pier and footing to resist the thrust which would lead to a large footing? I have designed MB foundations in the past with pier/footing resisting designs and the contractor always complains about the footing size. Recently I have been using a strip footing between the piers to counterract the force.

With this application, with a basement, I am trying to take advantage of the passing the load through the floor system or with independant beams.