sandwich existing wood beam with steel plates

huangyhg

sandwich existing wood beam with steel plates?
hi all,
i have a situation where there is an existing continuous, simply-supported 4"x12" (3.5"x11.25" field measured) wood beam. there is a column at approximately midspan. the beam is exposed (painted white to match ceiling and walls), and the owner wants the column removed because it interferes with the "flow" of the area. the beam supports a flat roof (2x6 rafters at 24" oc with 1/2" plywood sheathing, asphalt shingles, and 1/2" gypboard ceiling).
using 10psf for dead load and 20 psf for live load and assuming the existing beam is made of douglas fir-larch #1 with a span (ctr of bearing to ctr of brg) of 24.6 feet and a tributary loading width of about 8 feet, i have determined that i can remove the column if i sandwich the existing 4x12 with either 2-7/8" x 11-7/8" trus joist 2.0e paralam psls or 1/2" x 11-1/4" steel plates on each side. the psls should keep the dead load deflection under 0.4" and the live load deflection under 0.6". i think the steel plates provide about the same, if not better (i don't have my calcs in front of me).
does anyone have any opinions on what is the "better" option (wood or steel), regarding both common engineering practice and constructability (i.e., cost)? the owner, of course, would like the retrofit to be as thin as possible. also, any advise regarding providing the additional steel or psls on the same side of the existing beam for aesthetic reasons, e.g., using a single 1" thick steel plate or a thicker piece of manufactured wood on one side only?
i've been designing concrete bridges for the past 10 years so i'm a little bit rusty on the latest in residential or light construction, as are all the other guys in my bridge design office. so, you're probably wondering why i'm doing this. well, my mom asked me to so i felt obligated to look into it for her. if you think i should contact another engineer who specializes in this kind of stuff, please don't hesitate to let me know. the calcs seem pretty straightforward to me, though.
find a job or post a job opening
where are you located?  are there any possible drift conditions to consider for snow?
i would think that having a plate or psl on both sides would look better than on one side (more symmetrical), but hey that is why i am just an engineer, not an architect.
have you considered ponding for this flat roof?  you should definitely investigate ponding for a flat roof with a longer span like this.
offhand, sounds pretty skinny in depth.  you don't say if the beam is supporting a second floor or if its a roof.  if its a second floor, with the minimal depth you plan to use, i would think you will have vibration problems that would be very uncomfortable for any of the upstairs occupants.
with the longer span, lateral bracing of the bottom chord for tension collapse or turnover would become important, negating the asthetic benefits of the flat face look.  
what's the new span's end bearing surface look like.  watch out for crushing perpendicular loaded wooden   
biginch:
what is tension collapse?
blastresistant:
in my area (nebraska) steel plates are cheaper than psl. which always surprises me a little. but i have had more than one contractor tell me this, so there must be something to it.
i, personally, would put plates or psl's on both sides. just something about symetry of loading that makes me feel better.  
surprised that the sizes you have calculated for the psl and steel options are so close. especially in deflection; as the e of steel is so much greater than the e of the psl.
i agree with structuraleit, make sure you consider ponding in your design.
the bolts work better with the timber if there are plates both sides. i would use plates both sides if i could.
thanks for all your responses thus far. to answer your questions without long drawn-out explanations:
- ponding is not an issue,
- snow loads are not relevant (honolulu, hawaii),
- no 2nd story, just roof load (asphalt shingles, 1/2" sheathing, 2x6 rafters at 24", 1/2 gyp bd ceiling),
- the "new" end bearing condition will be the same as the existing. the owner wants to avoid opening up the walls, so the new members will not be bearing on the existing supports (assuming 4x wood posts). (this is not a problem, is it??) i am checking the fc_perp of the existing beam based on the existing conditions (without the supporting intermediate column and with the additional weight of the new members). it works out to be ok - about 395 psi < 625 psi (per nds 2001). shear also works out ok - about 120 psi < fv).
also, i'll be sure to triple check all my calcs regarding sizing of the additional   
i would respectfully suggest checking your roof structure for net uplift due to wind (0.6d+w) in a 105 mph typhoon prone area.  this net uplift value will be on the order of your 20 psf live load down depending on location, trib width, etc.  
the old beam and tensile load path probably does not "work" numerically as it stands, but you have changed the forces in excess of 5% and technically need to make this structural modification meet new codes.  in eliminating the center post, you are increasing the tensile uplift load on the remaining outer columns by roughly 270%.
sorry about the "tension collapse" term.  i couldn't think of anything better to call it.  if the tension chord is not laterally braced and a vertical deflection of the truss results in a lateral deflection of the chord, that chord eventually pulls up and lays over flat.  is there some other word for that?
   getting two 24'-6"long 1/2" steel plates is propably too difficut, you would most likly need to weld a few shorter plates together. you should consider 12 inch channels back to back, either side of the beam. trying to strenghten the beam without bringing the new plies to proper bearing is not a good idea
refer the other posting on flitch beams in this forum.