bottom flange bearing on steel beam

huangyhg

bottom flange bearing on steel beam?
i have a situation where a residential framer has connected 16'-0" 2x10 spf #2 floor joists @ 16" o.c. on both sides of a w10x26 a-992/50 i-beam bearing on the w10x26 bottom flange only. the 2x10's butt up to the i-beams vertical web. my strucual engineering software will not calulate this unorthodox loading. i believe additional connection details will have to be made. any advice would be greatly appreciated.
before i give my response, how did you post the picture here? i have been wondering for long time if this is possible. i would appreciate your input on that.
i have few comments regarding the design:
1. how are the ends of the 2x10 connected? are they just sitting on the flange and the osb?
2. why do you call this unorthodox loading? it is no different from the steel beam being loaded on the top flange?
3. why are you using software to design on this? you can if the steel beam is simple span, run simple wl^2/8 and get the beindgi moment and check on flexure (s req'd = mactual*12/fb) and make sure sx provided is larger than the s required.
4. from stability point of view, i am more concerned about the end connections of the 2x10s
simpson strong tie makes joist hangers that work with steel wide flange beams. i suggest that you use them or at least check them out. i can think of hhb type connector. here is a link to their web site with section and load carrying capacities:
amen, "lutfi"...
nice post and five gold stars for the graphic...how did you accomplish this?
thanks for the quick response.
i am not an engineer. i came to this forum for some professional help. i am an aibd professional   
here is an actual photograph of the situation....
along their length, it looks like the joists are "wedged" firmly into the beam. if this is true, blocking, probably 2x8s, between all the joists and the end walls sould keep everything stable. in other words, completely cover both sides of the steel beam web with 2x8s.
the only thing that comes to mind is that the top flange of the steel beam may not be considered braced at every joist location, so its unbraced length might be pretty long.  i can't say if it works or not of course, but other than that i see no problems with the detail except adding some blocking like sliderule said.  a longer unbraced length isn't a problem either as long as it's been accounted for.  
nice job posting pics *thumbs up*
sliderule, i agree with you and with ucfse as well. on similar details, i used to attach a 2x nailer flat on the fop flange and bolt it with 1/2 inch diameter bolts in staggered manner.
it is perfectly acceptable to bear on the bottom or top flange and the wb as well. as long as the load is transffered to the beam. you need to keep the joist ends as close to the web as you can to reduce the eccentricity (torsion). you need to meet stability requriements as well for the steel beam (the unbreaced length) and the joists as well.
i would add some blocking between joist ends, as sliderule stated. i cannot tell if you have any bridging between the joists along their span either. ouch, i am not sure if that meets code.  i think the top of the joists are braced by the flooring. however, the bottoms are not.
thanks for the hint. i am sure as engineers; we can do much better sketching than writing.
regards,
lutfi
isn't it normally considered to load the i-beam from it's top, transferring the loads through the web? if just the bottom flange (which is only 0.440" thick) is loaded, what is to keep the flange from separating from the web? wouldn't the i-beam have to be calculated with just the load applied to it's bottom flange?
the force required to shear off the bottom flange is about 5000# per inch depending on web thickness, so not a concern here.
jmhdesign,
this is why i said above that you need to consider torsion. in ideal situations, the load would be applied directly through the shear center as you have drawn. however, in real life that is not possible all the time. even if steel beams frame into the w10x26 webs, they still have eccentricities that would cause torsion (or biaxial bending). most engineers ignore these moments if they are too small. however, they can creep up on you can be cause significant stress increases. in my practice i avoid torsion as much as i can by preventing it.
regarding your question about what keeps the flange from separating from the web, well good observation. you need to consider your loads and make sure this does not happen. the code dictates how much stress you are allowed on steel beam and they leave enough margins that is called safety factor.
the flange to web zone will see the following loads/stresses:
1.    direct vertical shear
2.    flexural stress due to bending moment from 2x10 bending moment
3.    torsional stress if one side exerts higher load than the other side
you need to combine the effects of the above stresses and determine there maximum stress in the zone/area.  
this application is not unique to your situations. i have designed many overhead cranes and monorails. the trolley wheels travel on the beam flanges and load them in identical manner as yours are loaded. you need to keep the actual stresses at or below the code allowable stresses. as n engineer, i have made judgment call in the past where i exceeded the allowable stress by as much as 10%. i did that based on the following:
1. my computed loads are based on theoretical maximum loads.
2. i had knowledge in real life that the probability of reaching/achieving these maximum loads is highly unlikely.
now i like to comments about your use of software:
1. be careful using software. you need to understand how it works, its accuracy and most impotently you need to know the governing codes. you also need to be able to interpret the input and output parameters. i can give you simple example regarding unbraced length of the members that you have in your example (steel and 2x10). what did you use? on the 2x10 did you use repetitive