Deflection Limits 2

[dougantholz]

I have a question, and I hope that some of you out there may be able to give me some guidence. I am relatively new to structural design, I have only been doing it for 3 1/2 years. Everyone I talk to seems to give me different answers on the servicability limits on vertical beam deflections. I am wondering if there is a published resource that I can use consistanty,

ASCE 7-95 lists the basic servicability check for beams used to support a floor at L/360 for live load. It also uses 1.5-inches as a maximum for dead + 1/2 live.

Ignoring vibrations as a servicability check, a 35-foot long beam would have a maximum live load deflection, using this criteria, of 1.17 inches. The maximum dead + half live load is 1.5 inches.

If then we take T. Galambos's paper from 1986 (published in ASCE Journal of Structural Engineering) the probability of seeing the actual deflection is 80-99 %. This would seemingly strengthen my confidence that my servicability calculation is a real number and the beam would actually deflect that much.

Many experienced designers and project managers in my office seem to cap the deflection for live loads at 3/4" for an interior span and 0.3-inches for spandral spans carrying brittle cladding (the latter is taken from ACI-530).

In the commentary of ASCE-7 it states that deflections of greater than 3/8-inch would damage interior partitions, regardless of the span of the beam.

In reality, what should I be concerned with? Should my deflections be capped at 3/8? 3/4? 1-1/8? or should I follow L/360 without concern?
I have a similar problem with roofs, if anyone has additional insight.
Thanks
-Doug

 

[phuduhudu]

It does completely depend on what your beam is supporting, hence the many rules of thumb. A plain beam can deflect as much as it wants to if it has nothing on it that will crack. Of course people don't like to see things deflect and I believe span/360 is about the limit of what you can detect with the eye. But even this can vary if there is another horizontal line close to the beam that you can use as a reference.

For limits I generally use span/360 for brittle finishes like masonry and span/180 for flexible finishes. I believe these are the limits in BS5950, the British Steel Code, though I don't have it to hand. I believe engineering judgement is very important with selecting deflection limits and you should always ensure you select the appropriate one from codes. Carl Bauer

http://www.bauerconsultbotswana.com

 

[chandr]

Should also take a look here:

AISC Steel Design Guide 3: Serviceability Design Considerations for Low-Rise Buildings.

and

Eurocode 3: chapter 4. Mine is ENV 1993-1-1, which is quite old. Would appreciate if someone can point me where/how to get a newer.

 

[LPPE]

For masonry, I usually use l/600.

 

[dougantholz]

ACI 530-99 caps the L/600 at 0.3 inches now. But what about interior spans? Most slip tracks I have seen allow for about 1" of deflection at the absolute most. After than the load path changes. Is this a real problem or only on the calc pad?
Thanks for the responses.

 

[Trypod2]

If you have a masonry veneer, then the maximum deflection should not exceed 3/8 of an inch. Never more than the mortar joint width. Any greater deflection will create hairline cracks along the joint where it touches the masonry unit.

 

[LPPE]

Dougantholz - It seems that deflection limits are up to the engineer who is designing, and not to exceed the limits set forth in code (l/240, l/360, l/600, etc.). Lets just say an l/360 delf. limit for a particular beam comes out to be 1.45". I personally try to keep deflection for any beam under 1". So I would use a stiffer beam, but I could go lighter. I think it comes down to what you feel comfortable with designing for your client, as long as you do not exceed code limits.