CFS / Wood Structure Interior Bearing Walls on Thickened Slab 2

[RFreund]

In regards to say 3-5 story cold formed steel or wood framed structures for condo/hotel/multi-family type structures. Does anyone use thickened slabs for interior bearing walls? How about if they also act as shear walls?

In my particular case all the foundations will be on vibratory stone columns.

My concerns are:
Uplift resistance at shear walls.
Differential settlement where the thickened slab meets the exterior wall.

We are working with a contractor who has stated that they are using thickened slabs for a couple 5 story structures currently. I suppose the differential settlement may not be an issue because of the vibratory stone columns. Also I could size the thickened slab for bearing pressure and possible check to see what is required for uplift. Although uplift seems like a long shot...

Thanks

EIT

http://www.HowToEngineer.com

 

[jayrod12]

Typically your strap brace is fastened to each stud it crosses no? It's almost like sheathing, I would likely account for the dead load as normal (but an "accurate" dead load probably in the range of 7-10PSF for walls and 15 PSF for floors and roof depending on construction of those).

 

[RFreund]

Typically your strap brace is fastened to each stud it crosses no?

True and I started thinking about this. I suppose if you design the fasteners such that you could lift the trib weight of the stud, then maybe you could consider it. It would put the stud in bending, but there is a load path there that you could try and follow. Most likely you will have some sort of wall covering to assist as well.

I would likely account for the dead load as normal

Normal meaning as you would with a sheathed shear wall i.e. assuming rigid body or assuming half wall length as trib?

Thanks!

EIT

http://www.HowToEngineer.com

 

[KootK]

They are all screwed but, if I understand RFreund's concerncorrectly, it won't be nearly enough to get the job done. I only showed one brace below because it seemed more illustrative that way. Same idea with two.

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.

 

[jayrod12]

I guess I'm risky on this one then. I just can't see a wall lifting off of the floor without it having to pull on each stud. Yes it puts each stud into weak axis bending, but I still see there being a reaction at each stud along the centroid of the brace.

How to quantify it? good question, that's why I don't like designing steel studs. That's typically a by others in my book. I'm not well versed enough in it.

 

[KootK]

I'd never considered this until RF pointed it out here. Even if the screws could transfer the load to the braces, ther's no way in hell the braces could handle the induced flexure.

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.

 

[jayrod12]

I'm not sure where the flexure comes from in any substantial form. I still consider the load acting along the brace only.

I think your Koot brain is in overdrive today. Your sketch does should there will be flexure but I would consider it nominal and think of the brace as a tension only item. Sort of like a truss I guess.

 

[KootK]

Definitely on overdrive. Just put together a fee for $130K worth of ill-defined renovation. It's a terrifying amount of risk and I've got adrenaline coursing around all over the place. So, if the "P" loads are not suspended from the brace, what holds them up? It's not the sheathing because it's not that kind of wall. And it's not the ground because we've counted on the loads riding along with the wall when it tries to over turn.





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.

 

[jayrod12]

Why can't the uplift partially being resisted by dead load transfer take place in the stud? Again, similar to a truss joint. Maybe I'm visualizing this problem incorrectly.

The brace pulls along it's axis. At each stud there is going to be some reaction that will have both axial and lateral components. This axial load becomes slightly offset by the dead load carried by the stud. And a chunk of the lateral would be share across multiple studs through blocking and the weak axis capacity of the stud.

Again, I may be completely in the wrong here. And provided they are putting any sort of sheet finish on the wall (drywall, osb, sheet metal siding) there is going to be some form of diaphragm stiffness that will force the wall to work as a rigid unit.

 

[KootK]

When I draw a free body diagram of a "P" loaded stud post being lifted of the ground to help with overturning, I only see three options for balancing "P".

1) Suspending the post from the top track acting as a beam spanning to the shear wall chords. Fat chance.
2) Supporting the post on the bottom track acting as a beam spanning to the shear wall chords. Fat chance.
3) Transferring the post's axial loads into diaphragm shear. However, unless the sheathing and its fastening have been designed to transfer the post axial load out to the shear wall chords, there's no reason to assume that this would be feasible.
4) Axial load in the bracing providing a vertical reaction at the intersection with the studs. A couple of screws might do for a uniform load but not for a serious concentrated load. Also, as you mentioned, for the diagonal bracing to provide a vertical reaction, you've either got to deal with the horizontal component as:

a)Weak axis stud bending which seems unlikely/undesirable.
b)A horizontal reaction through blocking and into the sheathing. This would require blocking at every stud/brace crossing and, again, intentionally designed sheathing and fastening.

5) Flexure in the brace. Fat chance.

I may need a sketch of the mechanism that you're seeing.

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.

 

[RFreund]

KootK -> thanks for not letting this die!

Below is an image of the "tension load path" that I think Jayrod is describing (or atleast I am envisioning). This has sparked my to model at strapped wall using Ram Elements.

Forces

I have started to set up the model. I used a spring at the bottom of the tension chord for the tension hold down and also put compression springs at the top of the wall studs to simulate dead load (although I suppose I can actually load the wall with dead load). The hard part will be how to model the node for the stud/strap intersection (meaning both fixed so relative rotation cannot occur, or to pin one of the members). Thus far it seems like weak axis bending of the stud is governing.

In the meantime I had reached out to Cold-Formed Steel Engineers Institute and asked them what they thought was common or logical. Their response was:
"You are correct there is some debate, but I believe the tributary load to the stud is the most logical."

A couple things to keep in mind:
1. It seems to me that half the dead load (or rigid body) has been used before.
2. It seems that there has been adequate performance in the past.
3. Neither of the above arguments makes it correct.
4. I'm sure there are also other factors "helping" the LFRS system (i.e. partitions) similar to residential designs.




EIT

http://www.HowToEngineer.com

 

[KootK]

Well, I had to keep this alive. I steered you wrong with my previous response. And I've been been doing it incorrectly myself.

I see the path that you and Jayrod are proposing. I just question it's strength and stiffness capabilities. I think that you could make do with a much simpler model. Basically just my sketch with supports at the boundary studs only. If the interior studs are to assist with OT, they'll need to be able to make due without any manner of vertical support at the bottom.

Thanks for sharing the results of your CFSEI conversation.

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.

 

[GerryBertier]

The wall is almost always sheathed over with gyp board. If it was a gyp board shear wall, we'd consider it a rigid and use half the length.

Interesting exercise but not something I'm worried about.

As for the topic of the thread. I've done many thickened slab interior foundations for party wall bearing hotels without any immediate issues. I have been called out recently by a geotechnical engineer asking how I deal with the differential settlement. I didn't have a good response for him other than the "that's the way we have always done it." Not good.

I design them for allowable bearing also. I know there is the Army method which uses subgrade modulus but I would have 7 foot width thickened slabs at 13 foot on center.

For these interior thickened slabs... How are you designing them? By allowable? By modulus? What edge slope? 2:1? Where are you putting the saw joints? Centered on the spacing or at the thin edge?

I am interested in how your model turns out R Freund.

 

[RFreund]

OK, so for some further information.
CFSEI has provided some further information.
A reference to Structural Design of Low-Rise Buildings in Cold formed steel, Reinf. Maosnry and Structural timber by J.R. Ubejd Mujagic, et el. here chapter 6. I have the portion of text that I will try to post as well. I don't think that would be breaking any international treaties, etc.
The authors discus the lack of guidance and offer 3 suggestions. The discussion is actually in regards to to sheathed walls but CFSEI suggested the ideas applied to both sheathed and strapped walls.
1. Neglect the dead load
2. Rigid body (although 3 usually shows that this is unconservative for uplift)
3. Beam on elastic foundation. - They say that the rim joist along with the top/bottom plate above acts as a beam (so EI of the rim joist plus 3 top plates). The double top plate is a 3" spring at each stud. Then you can find the length of decay based on a Heteni (not the sword maker) equation. Usually you will find that this decay length is on the order of the stud spacing (again making #2 a bad assumption). However I wonder if you could get a sheathed wall to calc out like a box beam (top / bot plate are flanges), then you'd have a large EI. However I'm not sure the best way determine the EI of a strapped wall.

CFSEI also let me in on a DRAFT paragraph for an upcoming release of AISI D113 Shear Wall Design Guide.
This is subject to change!
(Type I = segmented shear wall)
(Type II = perforated shear wall)
1.6 OVERTURNING RESTRAINT
AISI S213-07 requires that overturning restraint devices (hold-downs) be used at each end of a Type I shear
wall segment and at the ends of a Type II shear wall. It also states in Section B3 that these hold-downs be
designed to “transmit the induced forces and, where required, the amplified seismic loads.” The amplified
seismic loads are determined using the applicable building code load combinations including the overstrength
factor, Ωo. While dead load may be used to resist the uplift force at Type I and Type II shear wall ends and the
uniform uplift force between Type II shear wall ends due to overturning, the engineer must be cautious in using
dead load that may not be transmitted to those areas of the wall experiencing these uplift forces.

One approach is to only consider dead load occurring immediately at or very close to the chord studs. It is not
appropriate to assume that the Type II shear wall is a rigid body and use dead load away from the ends of the
wall to reduce the overturning requirement at the wall ends. FEMA 451 [21] entitled NEHRP Recommended
Provisions: Design Examples wood framing Section 10.1.4.10.2 states “… calculations involving seismic
overturning and counter-balancing moments are assumed not to be applicable for perforated shear walls, as
they are not expected to act as rigid bodies resisting global overturning.” Wood framed perforated shear walls
are similar to Type II cold-formed steel framed shear walls and thus similar considerations should apply.

I setup my model but I still don't think it is a good model. I have a wall that is 10' tall x 9.33 feet long with 10Kip horizontal force. My first attempted was to put a tension spring at the bottom of the tension chord and compression springs above all the studs. This gives me an idea if the studs will be engaged and to what proportion of deformation, but something is not right with that. If I pin the bottom of each stud then there is some contribution to resist uplift (about 10%)... I'm tired though so I'm probably missing something.

EIT

http://www.HowToEngineer.com

 

[KootK]

I thought of another consideration. In a discretely braced wall, the shear load is not delivered uniformly to the bottom track. Rather, it is delivered to the node where the brace meets the floor level and then dragged into the bottom track. Unless the bottom track is a one piece member, the entire bottom track may not participate in resisting shear. Ever seen a bottom track tension splice? I haven't yet.

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.

 

[jfmann]

Interesting.........if, as appears the case.......a single tension strap is providing all bracing for this short shear wall, then (as noted by KookK in post above) you have what is a classic bracket assembly.
Wall studs are not going to be part of this bracket. For dimensions and force (10kips) listed above, reaction forces at low end of tension strap are 10.7 kips vertical (resisting uplift) and 10k horz.

First....is there really enough weld at top and bottom of strap to resist 14.7 kips tension force in strap?.......or 7.35 kips if there is another strap on far face of wall.

Dead Load of wall.....about 933 lbs total (if that)......does not resist much even for (liberally) full participation, which is not appropriate since wall would have to move upward way too much
to engage weight of wall (and then, only after tiedown failed or at least stretched).

For general issue of using slab to resist uplift.......relatively thin slab would have to be reinforced unless transverse width required (perpendicular to wall) is quite short.

For this particular condition......with 10,700 lbs uplift.........even if slab is 6-inches thick (at 145 pcf, unreinforced)........total area required is 148 square feet,
which is large enough, requiring a square 12 feet by 12 feet (that requires reinforcing).
If end of wall is at or near edge of slab (without connection to exterior foundation wall), there is obviously a problem to be resolved.

As for idea floated in one post........about resisting uplift force with weight of stone-pier (rod & steel plate at base of pier).......even if pier were 24 inches diameter,
for this case (10,700 lbs uplift) height (depth) of pier would have to be on the order of 34 feet without considering friction along sides of pier (which is speculative at best without testing).
Also, there may be relatively movement due to compression of stone when resisting large uplift force unless it is "precompressed".

John F Mann, PE

http://www.structural101.com

 

[RFreund]

John,

Thanks for the comments. A couple quick clarifications/questions (to all).
When I am referring to dead load, this would be the wall weight and any floor/joist framing bearing on the wall.

Is there really enough weld at top and bottom of strap to resist 14.7 kips tension force in strap

There better be! . But for real, there should be. Gusset plates can be added if required.

Does mesh count as reinforcement?? I mean if we are talking about an extreme event (not serviceability cracking). Maybe increase the width of the thickened slab at the chord elements...

EIT

http://www.HowToEngineer.com

 

[jfmann]

WWWF ("mesh") can work only if it provides adequate area of steel for required moment capacity.
To reinforce 6-foot long cantilever of 6-inch thick slab (due to self-weight).....with bars at mid-depth (to be realistic about placement capability and to handle positive and negative moments)
..... requires 0.16 in^2 per foot of Grade 60 reinforcing (#4 at 12 inches), for one-way bending.

Alternative to consider might be soil-anchor (helical pile).......which will support downward load and resist uplift.




John F Mann, PE

http://www.structural101.com