Parapet Above Concrete Beam

[jdgengineer]

I am designing a concrete spandrel beam with a parapet above. The beam is 18" x 28" and has an 8" thick concrete parapet wall above. From the bottom of the concrete beam to top of the parapet is 54".

Would you design this beam as a rectangular 18" x 28" concrete beam or would you design as an L-shaped beam including the parapet? We can handle the torsion design of the beam in either consideration. My preference would be to design as 18" x 28" but I'm afraid if I design it that way that the beam would be overly reinforced as the "top" steel would actually be "bottom" steel when including the parapet height.

Thoughts?

 

[MrHershey]

I typically just do the base beam unless it's either high seismic and I need to count everything that's actually there or the base beam rebar is getting significantly heavier than I'd like. Main thing in that case is dowels from base beam to parapet would need to be sized both for out of plane plus whatever shear friction transfer you would need to transfer between base beam and parapet. Similar to headed studs in composite floor design.

 

[jdgengineer]

Thanks Mr Hershey. What about the potential for an over-reinforced section? Have you typically checked this?

 

[jayrod12]

Do you need the top steel to keep your section under-reinforced. I would expect the added height should still ensure the bottom steel yields far in advance of brittle failure no?

 

[jdgengineer]

It's a condition that is bothering me a little bit so we have a bit of extra steel in it and are likely being very conservative in the design of the beam.

It's a small poolhouse that has a very heavy roof deck. The roof framing is steel & pandeck framing which will have a heavy tile roof deck and significant planters above.

The building length from end to end is 33'. There is a 20' pocket door in the front. On either side of the pocket door there are 8" thick concrete walls which are aligned with the outside of the 18" wide concrete beam.

We don't do a lot of concrete beam design so I'm trying to think of all the potential load paths and make sure our beam meets the design of all of them. I'm sure it's significantly overdesigned but it's only one beam and the cost of it is not really an issue. I just want to make sure there aren't any problems.

We have evaluated the concrete beam as:

1) Pin-pin beam spanning 33'-0". We've also designed the beam for the torsion force due to the reaction of the steel beams framing into the side of the 18" width. We currently have (6)-#8 bars top & bottom in the 18"x28" section and closed ties @ 4" oc. There are a few #4 "skin" bars to meet the additional longitudinal steel requirements for torsion detailing. The top bars wouldn't be necessary for this design.

2) Fix-fix beam spanning 33'-0" with the torsion included as well. This is not as significant of a design criteria. This led to some of the top steel.

3) Fix-fix beam spanning the 20'-0" opening including the torsion. This doesn't govern but wanted to make sure that the edge of the concrete wall could take the compression load from the beam.

4) Multi-span beam with supports at 0', ~7',26', and 33'. The 7' and 26' supports were analyzed to induce additional torsion into the beam due to the fact that they are not symmetric with the beam.

We are currently ignoring any contribution of the metal deck and concrete fill to brace the beam for torsion. I think there is probably another load path where the torsion could be a T-C couple between the concrete deck and the steel beams to resist the torque.

 

[KootK]

Great. One more thing that I'd never considered but will now bother me forever.

If the parapet can be sacrificed in an extreme event, I would argue that your L-shaped thing still satisfies the intent of staying under reinforced. Once the parapet started crushing in compression, I think that the neutral axis would chase it's way down into the primary beam.

At that point, your top steel would become top steel again and you'd be left with a beam that would fail by way of steel yielding before collapsing. In service, your over-reinforced pre-beam may be a little less ductile than one might like but, then, sometimes sacrifices have to be made in the interest of practicality.

Of course, if parapet failure means that a 7000 lb concrete cladding panel will fall 50 storeys to crush a bus below, then I suppose that's another kettle of fish.

We are currently ignoring any contribution of the metal deck and concrete fill to brace the beam for torsion. I think there is probably another load path where the torsion could be a T-C couple between the concrete deck and the steel beams to resist the torque.

They'll probably put slotted holes in the beam connection to provide for field tolerances. If this is a mechanism that you really want to count on, I might:

1) Weld the connection after erection bolting.
2) Ensure that the embed plate has a ductile pullout failure mode.

3) Fix-fix beam spanning the 20'-0" opening including the torsion. This doesn't govern but wanted to make sure that the edge of the concrete wall could take the compression load from the beam.

Can the walls and the connections between the beams and walls handle the out of plane bending induced by the beam torsion? I could see that being taxing, especially if you've only got one central mat of rebar in the wall.

Another alternative here might be to simply eliminate the beam and call the whole assembly a 12" wide perforated wall panel. It would detail and form simply and, surely, a 12" x 54" beam would do the job. Give some thought to LTB but, based on my precast spandrel work in the past, it should be fine.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[jdgengineer]

Thanks for the response KootK. The wall thickness limit is 8". I've tried to push for 10" but arch is pretty concerned about the thickness of the assembly (there is a pocket depth behind the wall as well).

The wall likely cannot take the moment induced by the eccentric reaction on the beam. Instead, I've assumed the wall takes only the axial force and that the beam takes the eccentric load via torsion. Does that sound reasonable? The wall is 8" thick with #5 bars @ 12" oc each way.

 

[KootK]

Instead, I've assumed the wall takes only the axial force and that the beam takes the eccentric load via torsion

But are the torsional beam end reactions not the source of the out of plane eccentric load on the wall?

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[jdgengineer]

I'm not sure if I understand. The torsion is a result of the steel beams framing into the side of the 18" concrete beam and the fact that the walls below are aligned with the outside edge of the beam. See attached sketch.

 

[KootK]

I get it now. You're taking the beam torsion out at the integrated pilasters shown in your last sketch. AOK,

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.