Two way slab deflection, ACI 318-08 2

[SamETABS]

Dear All,

As per commentary clause "ACI 318-08, R9.5 — Control of deflections" if you satisfy the minimum thickness requirements of clauses 9.5.3.1, 9.5.3.2, and 9.5.3.3 there won't be any need to control deflections!!!

firstly, What sort of deflection it is referring to? immediate, long term?

secondly, I'm concerned about the long-term deflections and cracks! How can you justify not to check long-term deflection for a bay of 7.8m x 6m in a Lobby with 500x800dp beams all around the bay(LL: 5 KPa, SDL: 6 KPa)with just 180 mm slab thickness which satisfies the minimum slab thickness requirements of ACI 318-08, Clause 9.5.3.3 (For slabs with beams spanning between the supports on all sides)?

I'd appreciate to hear your thoughts.
many thanks.

 

[dik]

With beams all around as you have and a 7" slab and a 20' span, I wouldn't be too concerned with deflection. Design it as a 2 way slab supported on 2 sides.

Dik

 

[dik]

For deflections, you may consider calculating the deflection of an elastic plate fixed on all four sides and, maybe, triple this to give you an idea of the long term deflection.

You might consider cambering the slab in the midde by approx 3/4" to 1". The 1" camber will likely provide a bit of a crown to the finished slab.

Dik

 

[rapt]

SamEtabs,

You will never know unless you do the calculations. Deemed to comply solutions like this in codes tell you nothing about actual performance so you do not know what the deflection will actually be.

Dik,

Multiplying by 6 will give you a better idea, 3 for long term and 2 for cracking.
Cambering does not help if there are worries about effects of deflection of something supported by the floor.

 

[dik]

Thanks, Rapt... could be, but for a 7" slab with those proportions I had assumed that cracking as well as long term deflection would be minimal.

In any event deflection should be small...

Dik

 

[rapt]

dik,

You are a braver man than I. It is not a standard lightly loaded slab.

It has 6KPa SDL + 5KPa LL. It may be an end span or even a simply supported span. We do not know. How can you make comments like you have?

Sounds like it is Middle East with a heavy topping and therefore brittle finishes and possibly brickwork .

By my quick calculations, it is cracked and would fail in deflections, both visual and for brittle finishes.

SamEtabs,
If you have a slab where you are worried about the consequences of deflections, ignore deemed to comply solutions and do the calculations properly, allowing correctly for cracking and long term effects. At least you can then get a reasonable idea of what is happening, though it will never be 100% accurate, at least you will be in the ballpark!

 

[dik]

6x is OK...

 

[rowingengineer]

The standard says minimum thickness, not "the thickness shall be...".

Given this I agree with rapt, you need to calculate the deflections with the high loads you have suggested. I wouldn't be surprised if you need a deeper slab.

as for the x6 multiplying of the deflection, lets just hope it isn't post tensioned or similar, because that x6 or x3, will be pissing in the wind.

I fail to understand with all the work that has been done on long term deflections by Ian Gilbert et al, why people are reluctant to do the calculations for long term deflections. They don't take long if you have the right calc's setup.

The span on depths are good for checking and making sure you have a minimum ball park but other than that, they shouldn't be a substitution for full engineering calcs.

http://www.nceng.com.au/

"A safe structure will be the one whose weakest link is never overloaded by the greatest force to which the structure is subjected” Petroski 1992

 

[dcarr82775]

For two way slabs I always use a multiplier of 4-5 for long term deflections(after including cracking). How you handle cracking is a crapshoot as there are all sorts of methods out there. When my calcs match what was measured in the field I tend to think it is more by chance than skill.

I have seen all sorts of problems with slab meeting the 318 limitations. Typically when higher end/brittle finishes with tight tolerances are required.

 

[rowingengineer]

4-5 has been suggest by some experiments for normally loaded slabs in normal conditions, but once you move outside these boundaries, calculations are a must. I find my calculations match very similar to published results.

http://www.nceng.com.au/

"A safe structure will be the one whose weakest link is never overloaded by the greatest force to which the structure is subjected” Petroski 1992

 

[slickdeals]

@RE:
Could you please post a reference for the literature that you have used in your computations for long term deflections? This is certainly an interesting topic. There are many parallel threads on this forum that address this particular issue, one that seems to crop up over and over........just like shear flow!

 

[dcarr82775]

Rowing,

I agree, higher loads require a more detailed approach. TO me that is in no small part due to the cracking.

What do you mean by published results? Often I am trying to determine why there is a certain amount of deflection existing in a slab. In that context I have not seen repeatable calculations that consistently predict actual deflections.

 

[rapt]

Slickdeals/dcarr82775 ,

Search for papers by R I Gilbert on this. There are many.

BS8110 Part 2 methodology is reasonable good.

If you want a 'black box" solution (well really not black, probably more Grey than Black!), RAPT software as RE has aluded to previously.

But noone can give an exact solution. material properties vary, loading varies etc. For lightly loaded members on the point of cracking, there is a grey area as to whether it is cracked or uncracked and calculations either side of the line will vary significantly, but so will slab performance.

EG in a slab test Gilbert reported in ACI Journals in early 2000's, the initial deflection was about 3mm as the slab was uncracked. After 3 days it was about 9-10mm, because the slab cracked due to shrinkage restraint from the reinforcement. After 273 days deflection was about 29.5mm due to increased cracking and shrinkage and creep (load was constant). ACI code predicted 9.3mm deflction after 273 days! RAPT predicted 30.5mm.

 

[rowingengineer]

slickdeals,
In the thread below two articles are posted discussion deflection, one by RAPT and the other by Doug (IDS), these both explain the theory of the long term deflection.

thread507-299616

I also have attached an article by Ian that has some calcs for long term deflection, which I think would give you what you were hoping for.

Dcarr82775

I use RAPT for for a good upper bound estimate of deflection taking into account long term deflection is RAPT, Note: RAPT doesn't take into account shrinkage restraint due to columns and shear walls, but does take into account shrinkage reo restraint.

The FEA programs I am familiar with are slowly getting to this level, however most still do not take into account reo shrinkage restraint correctly in my opinion.

I will post an article by Ian Gilbert that has long term results for deflection as soon as I can find it. I found it to be a good way of testing the FEA programs about 3 years ago when I was comparing software.

http://www.nceng.com.au/

"A safe structure will be the one whose weakest link is never overloaded by the greatest force to which the structure is subjected” Petroski 1992

 

[demayeng]

RE

That is a good document you uploaded.
I like how he says in the start that measured deflections of identical slabs in the field after 1 year differed by up to 100% in some cases!

I've seen mention of shrinkage restraint due to fixed items (like columns) recently - where can I find out more about this? Does Ian Gilbert talk about it anywhere?


ps I heard NCE has (or had) Inducta Slabs?

 

[rapt]

Demayeng,

Gilbert's work is mainly on cracking, shrinkage and creep effects on cross-sections and members.

The external restraint effects are basic engineering. The slab shortens, the restraints resist shortening, inducing tension stresses and thus earlier cracking and higher steel stresses and more deflection. He has done some work on crack control due to these effects, but their effects on deflections are basically induced tension stresses. You can determine these from analysis.

 

[rowingengineer]

I have red flagged my post above after it was pointed out to me that the attached material is copy righted, thus management will remove the link. below is an open source document that gives similar information, if not as detailed. I will also contact Ian about the above paper.

http://ww2.integer.it/Web_1/database_locale/Fib Praga_2011/pdf/030 596 065 1 Gilbert SE.pdf

Dcarr82775,
here is a good paper for deflection results.

http://www2.civeng.unsw.edu.au/fileadmin/public/research/publications/uniciv/R-407.pdf

some more good papers from Ian Gilbert,

http://www.civeng.unsw.edu.au/staff/ian_gilbert/publications

http://www.nceng.com.au/

"A safe structure will be the one whose weakest link is never overloaded by the greatest force to which the structure is subjected” Petroski 1992

 

[rowingengineer]

Demayeng
Yes I do know a fair bit about inducta slabs and there other programs.

The software developer and I have had many discussions about key areas of the software and failure to give appropriate information in there help manuals (I haven't checked in recent time, this comment is based on 2 years ago).

While the FEA model does now include long term calculations beyond that of Kcs, most FEA programs in my mind fair to handle reo shrinkage to a level that would be acceptable to me. This is in part failure of the product to allow you to superimpose your reo layout on the FEA model. This means that reinforcement restraint is never correctly included. I could go on, but lets just say that I use slabs for strip evaluation.

do you want to know anything else in regards of FEA?

http://www.nceng.com.au/

"A safe structure will be the one whose weakest link is never overloaded by the greatest force to which the structure is subjected” Petroski 1992

 

[demayeng]

RE
I still don't get it though. Are you saying that it is something you should calculate on top of the existing deflection formulas? (EC2 or otherwise)

I would have thought that the inclusion of shrinkage restraint caused by the reinforcement steel would cover that, as the mat of steel cannot shrink shorter than the distance between columns - does that make sense?

 

[demayeng]

Regarding Inducta Slabs, it is what I use. I find it is very fast to use, so am just trying to reconcile it's use in my mind!

I think Slabs has come a long way in the last 2 years, but I still don't think you can superimpose your reo layout.