concentrated loads on elevated concrete floor slab

[jmy61]

I am analyzing a 40 year old two-story structure that is to support new, heavier equipment on the 2nd level. In addition to the new equipment the 2nd level floor will also support rack storage and be subjected to wheel loads from forklifts.
The 2nd level floor system consists of a reinforced concrete slab supported by composite steel beams and composite steel girders that in turn are supported by wide flange steel columns. Below-grade reinforced concrete footings support the columns.
Can anyone direct me to a design/analysis guide for determining how to distribute concentrated loads and wheel loads on an elevated floor system? (I do not have access to finite element software).
Thanks in advance for any direction you can provide.

 

[SteelPE]

There are methods for calculating this type of load including one that used to be given in the USD catalog. There is also a spreadsheet that does this type of analysis. Use at your own risk:

http://www.calculatoredge.com/structural/deckslab.htm

 

[jmy61]

Thanks SteelPE. I am not familiar with the "USD catalog" that you reference. Can you provide a little more info. regarding the catalog (who publishes it, where I can get it, etc.). I have a number of references, but I am not familiar with that one.

 

[asixth]

It wouldn't be too hard to calculate the distribution by hand calculations. The system of primary and secondary composite beams is prominatly a one-way system, so the floor deflections, moments and shears and going to be a superposition of the slab span, secondary and primary beams.

 

[paddingtongreen]

There is a theorem (I don't know if it is official or not but it is logical) that if you have a viable load path, even if it is not the one that the load would choose (the stiffest one), the viable path will support the load even if the choice path cannot.

After checking punching shear, I would distribute the load through the slab for a width, the width of the bearing area plus twice the depth "d" of the slab. (The rationale for this is the distribution of the load from a baseplate, through a foundation, per ACI code). Then I would check if a beam-strip of that width would carry the load, if yes, then go with it, if no, then distribute it to a beam strip widening towards the beams or girder. I prefer to stay with an equilateral triangle shape, but will consider more if I see redundancies that help. If additional top reinforcing was provided across the composite beams and girders, I would consider the slabs continuous over the beam but am reluctant to do so with the girders (both beam ends want to rotate in opposite directions to the point that the concrete is forced to crack, I always put a saw cut along the centerline of the girder). It is just normal work from there on.

Alternatively, you could use yield line theory to get the loads to the beams.

Before I started to write, I looked to see if it had been done before and save me typing, I didn't find quite what I wanted but you might find this link a help:

http://www.bridgeforum.org/us-scann...vanced Assessment and Analysis Techniques.pdf

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[SteelPE]

Check out page29 and 33-34 of the CANAM steel deck catalog found here. At first glance it appears to be the same procedure found in the spreadsheet.... and the USD catalog.

http://www.canam-steeljoists.ws/

http://www/v4/epublica.nsf/va_pub/usdsteeldeck?opendocument&frame=ecanams

 

[Lion06]

Is this a slab on metal deck or a constant thickness slab? I have a paper for a constant thickness slab that gives an effective width of slab based on its location along the span.

 

[jdgengineer]

<quote> There is a theorem (I don't know if it is official or not but it is logical) that if you have a viable load path, even if it is not the one that the load would choose (the stiffest one), the viable path will support the load even if the choice path cannot. </quote>

I don't agree with this statement. Stiffness and strength are not necessarily correlated. Load is attracted by stiffness (law of physics). If the load path you choose is not the stiffest path it's possible the load may eventually go that route but not until significant failure of the stiffer load paths. Think of two springs in parallel. One significantly stiffer but weaker than the other one. The softer spring won't engage the full load until the failure of the stiffer one which may have significant ramifications.

It's like putting a moment frame on the same line as a brace frame but designing the moment frame for the loading and not the brace frame. The moment frame won't engage until the frame has failed which may Cause unacceptable damage.

 

[SteelPE]

I went to a seminar given by the AISC a few year ago. The speaker basically said the same exact thing that paddingtongreen stated. As long as you have a valid load path and that system is capable of supporting the load then the system will not fail..... now you may get a few cracks here and there, but the system will not fail.

Jdgengineer

I agree to a point. If you place a brace frame and a moment frame on the same line and then design the moment frame properly (accounting for strength and serviceability) then what is the problem. Using your example, if the system is designed so the weaker spring satisfies all requirements what is the problem.... almost like a belt and suspenders.

 

[paddingtongreen]

"I don't agree with this statement. Stiffness and strength are not necessarily correlated. Load is attracted by stiffness (law of physics). If the load path you choose is not the stiffest path it's possible the load may eventually go that route but not until significant failure of the stiffer load paths. Think of two springs in parallel. One significantly stiffer but weaker than the other one. The softer spring won't engage the full load until the failure of the stiffer one which may have significant ramifications."

I cannot say you are wrong, but you would have to be very obtuse and go far out of your way to find a practical case where real failure would have to occur before the flexible elements took over. Because we are dealing with qualified engineers, I would expect that they would see the awkwardness of a bad application and avoid it. In the OP's case, it is a slab, if one assumes a beam strip, simply supported when the ends are actually fixed, what is lost?

The principle was used for structural steel connections almost through the whole of the 20th Century, until more liberal methods were discovered or invented late on.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[jdgengineer]

As long as the stiffer element has the deformation capacity to go for the ride after it has failed then yes it's not a big deal necessarily. But it has to have the deformation capacity. What I intended to say is that yes it may not be a "failure" but it could result in damage that is unacceptable. Say it was an unreinforced masonry wall and a moment frame. The wall may fail laterally first while also compromising the gravity supporting system before the moment frame engages.

The time where I do think that theory is appropriate is when you are doing a seismic retrofit on a capacity based approach. If you incorrectly identify the yielding mechanism it's no big deal because you will be conservative and will be designing everything for a higher load then the system may actually see. In this case it's conservative. This is the approach of ASCE 41 linear static approach.

 

[paddingtongreen]

I meant to include that when we designed structures BC (before computers) we used Moment Distribution, Slope Deflection, Column Analogy, Semi-Graphical Integration etc. to find the primary load paths, ignoring secondary and tertiary paths, same difference, it is conceivable that a secondary path was stiffer than the primary. Again, a competent engineer should see it coming and do something about it.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[jmy61]

Thanks for all the input.
Lion06, yes the slab is a constant thickness of 5 inches.

 

[Lemony]

For those who are interested, the concept of load redistribution is addressed in ASCE/SEI 7-05 as 4.8 Reduction in Live Loads. This places rational limits on its use and indicates that it is only applicable to transient loading.

 

[Lion06]

Check out this paper. Hopefully, it is helpful.