Single Angle Flexural Design question...

[Hemi79]

Hi everyone, First let me say that I have been reading a lot of the posts in here regarding single angle design. I have looked into the steel manual and examples. I just would like if someone can help me clarify one thing. Let me explain, I am adding 2 single angles of equal legs to a girder which spans 46' in a building done in 1912 (Yeah the owners want to keep almost everything), now I don't know if this will work obviously but I want to see what my design results are. Now being that I will be adding bolts to connect both angles on each side of the girder, this is a 2nd floor by the way,and I have the joists connecting to the girders @ 18" O.C. I figure that I can design this angle as "continuous lateral-torsional restraint". Yes? So am I to use F10-2 or F10-3 equation and simply use my 18" as the "laterally unbraced length of a member"?

Thanks in advance.

 

[hokie66]

It sounds like from your description that you are adding the angles as flange reinforcement, although they are attached to the web. So they are essentially in tension, or maybe in compression, depending where they are being added relative to the depth of the existing girder. Although part of a flexural member, the angles themselves would be axially loaded, provided my interpretation is correct. Perhaps you can provide a sketch or a better word picture.

 

[JAE]

Where on the girder are the angles? Near the top (compression) flange? Just under the flanges? Bolted to the web?

A sketch might help.
Thanks.

 

[Hemi79]

Forgive my drawing capabilities....

 

[hokie66]

I thought this was a steel girder, but see now that you are still working on the grossly undersized wood beam in your earlier post. I don't hold much hope for your approach of adding steel angles, as the beam is still much too shallow...maybe big channels each side would be a starting point...or a central column.

 

[Hemi79]

Well I proposed a central column on each one of those girders and it was a hard NO from the owners and architect. Channels was my next approach, but I wanted to see my results on adding angles. If it were up to me I would start working on possible plate girders, but its not up to me and I just want to put as many options together as are probable to work. This has proved to be a difficult task for me. I appreciate the input.

 

[bookowski]

I'm guessing that those are not continuous 46ft 2x10's? So they are spliced at least once? That seems pretty optimistic... even with channels at 46ft. - but to get closer to your question I would think the angles are acting as tension elements more than flexure if you are trying to make them composite with the beam and locating them on the bottom.

Have you thought about making it a truss with the existing girder as top chord?

 

[hokie66]

The first thing you need to do is to assert who is in charge of the structure, which must be the structural engineer. Architects often want to do impossible things (refer to the current Skyhooks thread in this forum), but we engineers deal in reality. My starting point would be to accept that the existing beams have failed, as there is no possibility of them being adequate in either strength or serviceabily, notwithstanding their age. And my second decision would be that the wood beams are to be ignored in the flexural capacity of the reinforced floor.

 

[dhengr]

If the Arch. wants to keep that beam, tell him you’ll cut it up into 8' length and pile it on his drafting table. That will get it out of this bldg. and maybe prevent him from drawing stupid architect things. Design a beam or truss which will actually carry the load, maybe add a column or two and you might actually be able to save the existing joists. Someone who knows what they are doing and has some Structural Engineering experience and judgement has to take charge of this job, and tell the Arch. and owner what is practically possible, irrespective of their dreams and wishes.

 

[Hemi79]

Bookowski: Yes the girder is spliced at least once. I will look more into the problem to see if I can see it your way regarding the tension vs. bending. I was hoping more for the angles to just take over most of the needed capacity. At first I wanted to set the angle on top, but for construction reasons I figured on bottom. In our meeting we discussed making the girder a truss, adding a beam underneath, and the channel/angle approach. I went with the channel/angle approach first. Im just not optimistic about adding any wood members for the 46' span. I really wanted to just make it steel.

hokie66: I couldn't agree with you more. I just want to have more calculations in hand before I tell my boss I don't see any other possible option but to add a vertical support.

Now, as to my original question regarding angles that are laterally supported. Is equation F10-4a or F10-4b x 1.25 appropriate using 18" O.C. for L? Being that the angle is laterally supported and not just @ midspan.

Thanks for the replies everyone!

 

[bookowski]

Have you run any initial numbers? I think you're going to find that the unbraced length is irrelevant - assume 0" unbraced length and see what kind of angle you'd need. You didn't specify what kind of loads you are carrying but even in a wf it would make sense to have a w18 or 21 for this span - an angle isn't going to cut it. Run some numbers.

 

[Triangled]

Agree with all said preceding. A second floor girder spanning 46', wood 2x10s? Sounds more like maybe a 1910 roof originally. what is your tributary width? Single angle flex is tricky. I wouldn't even look at it here, not because of trickiness but because it is self evident (to me at least) that it is a non viable solution for deflections for sure and almost for certain for bending due to the shallowness of the section. Good luck!
A quick confirmation for you if you wish, try running a channel .. U shape...flat, oriented as your drawing indicates, thus removing the dual axis asymmetry from the problem. I am sure you'll find it doesn't work. If that doesn't work, adding the single angle torsional item only makes it worse. I think boss will agree.

 

[PUEngineer]

Is your boss trying to punk you? Seems like she/he should already know, or at least have a sneaking suspicion, that a 10" deep wood girder won't span 46'.

Share with us your uniform design load for the girder. Maybe it's not really that bad...

 

[Archie264]

Hemi, are you sure that thing isn't the bottom chord of a timber truss? The bolt pattern in your sketch makes me think that it could be. Well, that and the fact that floor didn't collapse 100 years ago.

 

[BAretired]

I agree with Archie. The 6 - 2x10 members are probably the bottom chord of a truss. Check it out.

BA

 

[Hemi79]

Tirangled: Tributary width for joist is 18", spacing between girders is 20', girder span is 46'. The angle or channel didn't work. Not even close. Thanks for the input.

PUEngineer: Im pretty sure he wants to train train me. He never said he thought it would work, he simply asked me to put together solutions. UDL is 500 lb/ft.

Archie264: Its not a truss. I went in there and observed it. It was hot as hell and not much space to maneuver, but If this is a truss its the first one I see like this. Please see the attachment and tell me what you all think. Sorry for the bolt pattern, that's just something I put in to throw out an idea as how to bolt the angle to the girder, but the angle and channel are out of the question. The floor has deflection. Several inches in some areas, and when standing at the girder location it can most certainly be felt as to when standing over weak joists area. This I feel is mainly because every joist is notched with only 2" (out of the 10") bearing over the girder.

Going with a W21x55, so far it makes it. Gonna keep at it.

Thanks for tuning in.

 

[hokie66]

So those 6-2x10's are actually spaced as per your previous sketch?

thread507-355131

Sounds like you are on the right track now, with a big steel beam. If you only use a single beam at each "girder", then you will need headers to transfer the joist loads to the beam. And that connection of the joist to the "girder" is no good...install joist hangers. What a mess!

 

[Hemi79]

hokie66: Yes those are it. That drawing from my previous post I did to include in my report. Thanks for the advice on the headers and everything. I really appreciate all the help. Being that I solved for the inertia by inputting the allowable deflection my results are marginal, but flexure design is easily covered.

Thanks again everyone.

 

[Jerehmy]

I would remove the wood beams and replace with steel.

Wood-steel composites just don't work with that connection. Too much crushing at the bolt/nail connection to make composite action viable. Look up some of the experiments from the army regarding this (don't have time at the moment).

 

[BAretired]

The new beam will be narrower than the existing wood beam, so the joists will not reach far enough. How is that going to work?

BA