Stiffened Seat

[slickdeals]

Folks,
I have a rather stupid question to ask. I am trying to design a stiffened seat/bracket to support reactions from a beam.

The bracket will support a beam that forms part of an expansion joint. I am trying to use 3 vertical stiffeners in lieu of 1.

The reason is that I don't want to rely on bracked flange bending to resist the reaction. How can I design the bracket such that each stiffener gets its fair share (middle getting 1/2, other two getting 1/4).

How can I be sure that all my load is not going into the middle stiffener (one that is parallel with the beam web).

Thanks

 

[theonlynamenottaken]

Put another plate across the edge of the stiffeners, parallel to the plane of the web.

 

[kslee1000]

Pride a pair of bearing stiffeners to near the end of the beam, use thicker/rigid seating plate, or provide thick/rigid filler plate under the beam. This system now is similar to a continuous beam over 3 supports, the middle reaction will be higher than the outers.

 

[dougantholz]

Avoid this connection like it was the plague I've seen it fail quite often. That said why not a short section of wide flange. A lot less welding and it already has a stiffener, bottom flange and top flange for maximum welding to the column.
Ps. - don't forget to include the diameter of the bolt in calculating slotted holes. It often gets overlooked.

 

[slickdeals]

@doug
I like your idea. However, I would want to know reasons why you would avoid it like the plague. Just for my understanding.

 

[dougantholz]

Reasons - first I have actually seen failures. Reasons vary but failed connections should be avoided. Double columns is so much better if you can swing it.
My thoughts to consider:

Consider max thermal contraction and expansion in design. Load combination per ASCE 7: 1.2D+1.6S+1.6 Thermal.

Slot the hole at least 1" longer than you think you need.

Detail where to put the bolt based on the temperature during construction and temp expected during lifetime.

Carefully consider welds of seat to column, especially to a tube column.

Consider bending in face of tube column due to eccentricity of load.

Make sure your bolts don't have to transfer shear from lateral loads in a direction perpendicular to the member you are connecting.

Consider the stress in members perpendicular to the one under consideration. Especially as chords of the diaphragm.

That's the off the top of my head stuff.

 

[slickdeals]

Doug,
I don't think I really explained it properly. It is a seismic joint

The beam is bearing on your suggested stub beam, but it bears on slide bearings. The slide bearings have been sized for Cd * displacement using the SRSS of the 2 building displacements.

There is absolutely nothing restraining the movement of these beams.

Now, I will let you give me more reasons why you think I should not be using this stub cantilever.

 

[Lion06]

look at fabreeka or fluorogold slide bearings for expansion joints. We use them if at all possible at expansion joints.
Bolts in slotted holes are not overly effective expansion joints.

 

[hokie66]

Now that it is clear that you are using a slide bearing, back to your original question.

I don't know why you would want to have the centre stiffener take half the load. Why not have all three work? In the ultimate condition, they should all take the same load. And using bearing stiffeners in the beam and a thick cap plate on the seat will make that happen. Sure, the centre one will initially take more load, but it will then distribute.

 

[slickdeals]

@Hokie,
I am not sure how the load distribution can be achieved. I tried a EA/L spring model to see the load distribution and I could never get more than 10% of the loads out to the end stiffeners. I think the central stiffener that is in line with the beam web will carry the bulk of the load, till it deflects ever so slightly to redistribute the loads back to the other stiffeners.I would appreciate it if you could explain how I can distribute the load better.

Yes, I am using double bearing stiffeners in the beam in order to be certain that I will not have web buckling due to lack of bearing in the ultimate state.

I am however leaning on the idea of using a stub beam with web stiffeners, welded to the column flange (with horizontal stiffeners in the column), but the triple stiffener bracket plate was my original idea.

@STREIT: I am using a CON SERV CSA type bearing. We have used CON SERV successfully on many of our stadium projects before. Again, I reiterate that this beam is completely free to slide around (the slide bearing plate is sized big enough to let it move as much as it will in its ultimate state)

This is for a long mall (about 1500 ft long) which is 3 stories high in Trinidad.

 

[youngstructural]

Dougan: Can you expand on your comment? What occurs with the detail to make it fail?

I am understaning correctly that you're making a seat out of a length of angle with triangular pieces of plate welded into it, right. You say this often fails? How? Why?

Looking forward to your reply,
Cheers,

YS

B.Eng (Carleton)
Working in New Zealand, thinking of my snow covered home...

 

[hokie66]

slickdeals,

What I mean is that the centre stiffener will indeed "deflect ever so slightly" and redistribute the load to the others. If you want it to be determinate, just use either one or two stiffeners, and there will be no question of equal distribution.

 

[slickdeals]

Guys,
Could any one of you please review the attached calculations and comment on the procedure.

In addition, I would appreciate if you could answer the questions on that sheet?

Thanks

 

[nutte]

The AISC steel manual has a procedure for this type of connection. This procedure checks, among other things, local buckling. Have you checked it out?

 

[slickdeals]

Nutte,
I looked at the stiffened seat connection in the AISC manual, but I did not find anything related to buckling. Maybe I missed it.

 

[nutte]

Hmm, maybe you're right, maybe it's not in there. I know for certain it's in Salmon and Johnson's steel text.

 

[slickdeals]

I was following the Salmon Textbook, but because of limitations I have a fairly wide plate and the minimum thickness to prevent buckling is causing my plate to be really thick.

I am sure there must be ways to improve on the design. I am sure creative design solutions exist and that is what I am looking for in this forum.

Thanks for taking the time to look into this.

 

[nutte]

Yes, I understand your frustration. Unfortunately, I don't know of another way.

 

[youngstructural]

Why not check buckling in terms of a Johansson yield line analysis of the stiffening plate...

If you haven't considered this, I would be pleased to do the calculation and include references. Just let me know, as I would prefer not to waste twenty or thirty minutes for something you're ruled out.

Cheers,

YS

B.Eng (Carleton)
Working in New Zealand, thinking of my snow covered home...

 

[slickdeals]

youngstructural,
I have not ruled anything related to plate buckling out, in fact I would appreciate it if you could spend the said 20-30 mins to go over the procedure.

I am really concerned with the design steps and I feel that the only thing that I have not checked completely is plate local buckling.

Please feel free to tell me if there are some things missing that I should check or if it is complete.

Thanks