long span composite beam - advice requested

huangyhg

long span composite beam - advice requested
hi all.
i am designing an office floor over a very large conference room, where the span is almost 80 feet.  i would like to use composite beams for this span, but i am concerned about some of the sizes and deflections i am seeing.  using various calculations and tools, like the aisc bay studies software, i find i would need w30x108 beams to span across the room.  my loads are 63 psf dl, 25 psf sdl, and 65 psf ll (average office/corridor).  beams are spaced at 5.66 feet on center.  the deflections are dl=3.78", sdl=0.69", ll=1.26".  i figure on the camber being 2.5 inches.  i have never designed floor beams this long, nor with this high of camber.  also, the total magnitude of the sdl+ll deflection, which is total almost 2 inches, seems very high regardless of span ratio.  the beams also have to hold up moving partitions, so i am doubly concerned about the ll deflection affecting the moving partitions.
can anyone recommend any practical limits for camber depth, or maximum beam size/length which can be cambered, and absolute live load deflection limits?  or possibly point me towards some references for such long span composite systems?  also, are traditional vibration equations applicable to such long spans, considering that these beams have no back spans (they span from one side of building to the other)?
the only other option that i can think off here would be deep long span joists, like maybe 60" deep lh joists.  this way, the ductwork could pass through the joists (hopefully), and the ceiling could be tight under the joists.  
anyone have any other thoughts on floor system for these spans?
thanks very much,
andrew


deeper is better,  the joists you mentioned at a generous spacing, (say 15-20 feet), with w12 as intermediate beams between the joists.  for solid rolled shapes, w40x149 at 21 foot spacing with w12's in the orthagonal direction might be cost effective.  
80 feet with a w30x108 seems rather wimpy. i have done 63' parking garages (40 psf ll + 63 psf deck + 5 psf sdl) and ended up with w27x129 (partial composite)
have you considered vulcraft's composite joists?
quote:
80 feet with a w30x108 seems rather wimpy.
i'd second that.  usually you try for a depth of member equal to "span x 12 / 2" - so for an 80 ft span i'd use a 40" deep   
what size of deck and depth of concrete topping are you considering using?
for kicks, i plugged in your numbers, and assumed 3" deck with 3 1/2" normal weight topping (assuming this is where the 63 psf comes from) and i get a w30x116 using a post composite deflection limit of l/360. so, i think you are not all that far off. i used ramsbeam to run the numbers.
using a w36x150, i get a deflection of 1 1/8". this seems very reasonable to me for this long of spans. i have done 74' spans, and used w36's, so, a w36x150 would not seem unreasonable in this situation.
i think you are right to be concerned with the absolute magnitude of the deflections. especially considering the partitions. an l/360 deflection (2.66") would likely be too much for the partitions. i would contact the partition supplier (if known. otherwise any supplier) and ask what is the maximum post composite deflection the partition can tolerate without binding. this may be the controlling criteria. then, its just a matter of finding a beam that can meet the criteria.
  
be careful of annoying floor vibrations!!!  
an 80' span will inherently have a low natural frequency, even with deep beams.  this type of floor has a high probability of having perceptible floor vibration when people walk across the floor.
andrew gorton, msc
noise & vibration consultant
i agree w/ vibeac --- even if deflection is within limits - the vibrations will haunt you.  deeper the better - like use a joist.  
there is a very high end building here in st.l with a large meeting/wedding hall.  when people start dancing - it scares the cr&p out them....  structurally it is sound - checked by multiple engineers - just vibrates way too much.
nothing like drinking beer all night and then eating two triple stacks of flapjacks afterwards, then dancing on a trampoline with a 360 lb. friend
can you say, "hurl".

one way of getting greater depth without so much more steel is to expand the beams into castellated beams.  we used to do that a lot, but perhaps the fabrication costs would rule it out now.  same thing for purpose designed trusses, but the standard long span joist/trusses may be the way to go.  where i am, this would be a posttensioned concrete design.
another fairly effective, and inexpensive, way of reducing vibrations is to thicken the concrete deck.
i don't know if this is possible in your situation, as you may want to stick with a certain thickeness on the floor areas surrounding these long span beams, but going from say a 6 1/2" total thickness deck to a 7 1/2" total thickness deck makes quite a bit of difference in vibration charateristics.
thanks all.  glad to see others are just as concerned as i am about the deflections and vibrations.  my senior engineer is almost insisting on using "conventional" composite beams, but i am investigating other configurations to try to find the best solution.  
i really prefer to use my idea of the 60" dlh joists, even possibly make them composite, as i can get very low deflection (delta ll=1.0") and save a bunch of money.  my initial calcs tell me i could save about $200,000 compared to the w30 (or bigger) beams, when considered over the 4 floors that this is occuring on.  that is not an insignificant amount of money.
thanks for your support.  maybe i can show my senior the cost savings, and general opinions, and be able to change his mind.