Concrete design question 3

[AlmostPE]

I have not been exposed to concrete design too much in the real world and I have general questions:

1. If I have a structural slab that is suppoted by 4 walls. The slab sits on the top of the walls and dowelled into the wall with bent rebars (one leg into the slab and one leg into the wall). So is this fixed or pinned?

2. Do you guys have any pointers on how to design the slab? I was going to design it as if it were 1 way slab and design it to span the short direction and then put the rebars the same way both ways. Shall I follow ACI and make the slab thick enough so I dont have to worry about deflection? Some of you may say to ask my supervisor to train me but I dont think he knows any better than I do and he has a PE stamp!. :-(

Thanks!

 

[shin25]

1) You have to consider the joints as pinned connected, because, the walls may not have been designed to carry fixed end moments and the required amount of reinforcement for considering the joint fixed will be too large and may not be economical.

2) For deflection control, you can use ACI-318 suggested thickness.
For designing reinforcement, yield line theory is most applicable for this kind of situtation.

 

[BridgeGuy33]

1. From what information you have provided, it seems that the slab would be considered simply supported (pinned) by the walls.

2. ACI is likely the correct code document, but it would depend on where the structure is located and who has code juristiction. If you are in the US this should be the case.
You need to look at the ratio of the span lengths to determine if the slab is a one-way or a two-way slab. You'll need to evaluate how much deflection is acceptable and verify that you meet that requirement. ACI gives some guidance on minimum slab thickness.

Good Luck!

 

[AlmostPE]

Well, do you think my approach of designing it as one way slab (pin pin) on the short direction will be a satisfactory design? Then Ill just put the same rebar going the longer direction. This project is not very big so I do not have the time right now to really learn it to save amount of steel and concrete. I have a feeling the PE will love and trust whatever design I come up with (which is scary). So I want to make sure I do it right in so little of time.

 

[DonPhillips]

From what I remember of concrete design, you will have a different d in one direction compared with the other since the bars in one direction will likely sit atop the bars in the pther direction. This difference would need to be checked to ensure the slab can support the design loads.

Don Phillips

http://worthingtonengineering.com

 

[apsix]

Designing as a 1-way slab will be safe and conservative. The rebar spanning the short direction should be in the lower layer.
You should also have top steel at the supporting walls, especially at the corners, to avoid cracking due to negative bending moment. However, I would still design it for pinned supports.

 

[AlmostPE]

yeah, thanks apsix. That is what I was thinking too. Thank you everyone!

 

[oneintheeye]

you could design as two way (put the short direction steel in bottom) by use of tables. In the british standard theres a simple table to use not sure if equivilent in US. Should be considered simply supported. If you design as one way spanning design will be conservative so be ok, check deflection as one way spanning as well, (short span). effect of four support sides will mean deflection will be less than you allowed for. Generally if the ratio of spans > 2 then one way spanning.

 

[csd72]

Check out the ACI website for an example calculation.

http://www.cement.org/buildings/design_aids.asp

 

[vincentpa]

When dealing with a single slab bearing on 4 walls, you will also have to account for Mxy and Myx moments. You can look at PCA Design of Rectangular Tanks for the design procedure. I would definitely buy this book (it's only like $40) and use this design methodology. They have an example in the book for a tank design with a roof slab, which is basically what you are describing. Is the wall you are supporting below ground? If so, the walls will have to be designed like a concrete tank. They will not be braced until the top slab is poured. If you do not account for this in your design of the wall, you could have a failure or at least a wall with large deflections or large cracks. Also, be careful in your detailing of the wall. You should look in the ACI Detailing Manual for help in detailing a concrete wall or "tank".

 

[Lion06]

You could certainly design it as a one way slab and it would be safe (i.e. it won't collapse), but depending on the aspect ratio (I didn't see it mentioned above, just that one side is longer than the other. there is a difference between a ratio of long to short of 1.25 and 4) it really wants to behave as a two way slab. It won't collapse designing it as a one way slab, but you may have some serviceability issues.

 

[csd72]

AlmostPE,

These guys are just confusing you (they are almost confusing me and I have done heaps of this stuff!). Look up section 13 of ACI318 and it covers 2 way slabs. There is a chart that gives moments in each direction depending on the span.

These charts are based on yield line theory as mentioned by shin25.

csd

 

[vincentpa]

The Direct Design ACI Chapter 13 requires multiple bays. The Equivalent Frame method might be able to be adapted to a One-Bay situation but I don't know how accurate it will be. If you try to use the Equivalent Frame Method of Chapter 13, you will need PCA Notes on ACI 318 to get the factors to solve the problem. PCA Notes on ACI 318 also has a design example which is incorrect. The easiest solution is to use the method in PCA Rectangular Tanks. Or model it in a finite element or 3-d frame analysis program. I would use at least 8 plates elements per side (as a minimum) for a computer analysis or 1 to 2 foot wide plate elements.

 

[csd72]

Sorry I am looking at the 95 version and assumed the latest would be the the same.

 

[AlmostPE]

vincent.. hmm I do have PCA rectangular tanks. I didnt think about that. I will analyze with both pca rect tanks table and 1 way slab and see how close they are.

 

[twinnell]

I have also seen it done but setting the deflection equitions for each direction equal to each other to determine the load used for each direction (for simple beams in each direction). Essentially determines the ratio of load that is applied to each span direction. Has anyone ever seen it done this way?

 

[AlmostPE]

twinnell, I do remember learning something like that.


So here is what I come up with, the slab is about 16x22 ft. Based on L/20 I made the slab 10" thick.

DL = 10/12 X 150 = 125 psf
LL= 125 psf (IBC light storage)

Mu = 387.5 psf

I came up with #5 @ 10" o.c. bottom reinf. ea. way.
This is also good enough for shrinkage.
Now can I just stop right here and not worry about deflection since I used L/20?

Now my next question is, if I design this slab as pinned, can I use bent rebar into the wall? Wouldnt that create a fixed connection?

Since I designed it as pinned, how shall I dowel the slab into the wall? What design value shall I use since the moment is 0? Shall I just space it @ 10" o.c. also like the slab? Maybe embed it development length distance? I am just afraid doing this will create fixed connection then I have to redesign the walls to take moment. Sorry with all of these questions. I just wish I had a really good engineer working with me so I can ask him all of these questions.

 

[jechols]

I would put top steel around the perimeter and extend it into the slab to at least the quarter point maybe even the third point of the span. I would provide an L-bar that ties the negative steel (top) in the slab to the wall. As far as the bottom steel just extend straight into the wall. You might want to increase your dead load some, for ceilings, plumbing, mechanical units, etc.

J

 

[AlmostPE]

Im sorry.. I meant Mu=387.5 * 16^2 /8 = 13150 lbs ft.

 

[csd72]

I just realised what you guys are on about - ACI have dropped the 4 sided slab charts. Whats the story with that!?