shear tab with axial load 1

[rnorth]

how does AISC 13nth edition handle connections (simple two bolt shear tab) with combined shear and axial load?

in section 9-3 discussion is made about not combining stresses. user is refered to spec J to design connection elements.

would spec H be used for connection elements?

thanks in advance.....

 

[jberg]

I have asked that same question to AISC Solutions and they did not give me a straight answer for how to handle the interaction. So I use the concept posted in an article from AISC Journal, "Combined Shear and Tension Stress" - by Subhash C. Goel, 3rd Qtr.-1986.

Essentially, the reduction factor applied to the tension "limit state" capacity is = (1-R/Rv)^2.
where: R = actual shear end reaction
Rv = allowable shear capacity for the particular "limit state" considered.

Hope it helps.

JWB

 

[connectegr]

Yes you can combine shear and axial loads using single plate connections. The tabulated 13th Ed tables will not apply. There are several additional considerations. We design these connections frequently. This condition will also occur more frequently with the addition of tie forces.

Some considerations:
slots require slip critical bolts
bolt eccentricity
weld design (note this is not simply based on the end reaction)
weak axis bending of the plate
plate buckling
are stability plates necessary
adequacy of supporting member

http://www.FerrellEngineering.com

 

[jberg]

connectegr:

Good point about the weld design and slots requiring slip critical bolts. I am curious, how do you design for the combined tension and shear forces in the shear tab when a connection is subjected to an end shear reaction and a drag axial force?

JWB

 

[rnorth]

would you do a unity equation similar to spec H for combined loads? ASD omega is 1.67 or 2

Pa/Pn/omega + Va/Vn/omega + Ma/Mn/omega < 1.0

moment is from small eccentricty from bolts to face of column....
shear is from beam end rxn
Pa or Ta is from axial load

or would you analyize stresses with state of plane at a point from strength of materials

 

[connectegr]

Worst case: No lateral stability (no horizontal bracing or non-composite beam)

Shear Plate
plate stresses - shear, tension, bending strong-axis, bending weak-axis (combined stress interaction)

net fracture limit states - block shear for shear forces, block shear for tension forces, net capacity of plate for shear, tension, and bending

Beam Web
block shear, beam web tear-out

Bolts
plate/web bearing, bolt eccentricity (vertical and horizontal)

Weld %-)
IMHO - I check the welds for the actual shear and axial forces. But use at least 5/8 x plate thickness (upto 1/2" plate w = 5/16 max). For thick plates, design for the plastic moment capacity of the plate (I think this is conservative, but in accordance with AISC's intent)

Disclaimer...
This may not be the complete list of design considerations

http://www.FerrellEngineering.com

 

[nutte]

That weak-axis bending is going to result in a very thick plate. AISC has an example of a shear tab with axial load in the Seismic Design Manual, and it neglects weak-axis bending of the plate. In their example, there presumably is a diahragm that the beam is attached to, giving this weak-axis moment another path to be restrained.

 

[connectegr]

Yes, weak axis bending results in thick plates (squaring a fraction). That is one reason why single plates are not the best choice for axial loaded beams. But for skewed connections, column web connections between stiffener plates, or heavy framing where drilling thick material is costly single plates can be a solution.

If there is restraint in place, composite beam with studs or horizontal bracing, to resist out of plane movement, then weak axis bending would not apply.

http://www.FerrellEngineering.com

 

[DHKpeWI]

By Shear Tab do you mean the "Single Plate Connection" as shown on page 10-101 of AISC 13th Edition SCM?

If so it depends if the connection is considered to be "Conventional Configuration" or "Extended Configuration."

If you have the "Conventional Configuration" with "a" less than or equal to 3-1/2" and less than 10 bolts, eccentricity can be ignored on the bolts for the vertical shear. To include the axial load I would determine the resultant horizontal shear to each bolt taking into account any horizontal eccentricity.

With these two forces (vert. & horiz.) I would find the resultant force in each bolt and check bolt shear, bolt bearing, etc. Use Slip Critical Bolts if you specify SSL holes.

You could than apply the same priciple to the design of the "shear tab" (plate) and the welds of the plate to the supporting member.

Please see pages 10-101 to 10-104 for additional information and requirements.

 

[connectegr]

DHKpeWI,
By applying shear and axial to the single plate (shear tab) connection, the tabulated single plate connections and design method does not apply. This design (with no eccentricity) applies to shear only. This was based on lab tests of shear only single plate connections. The design required is similar to the extended plate design, and the eccentricities are relative.

And FYI, the single plate tables and design method will change again in the 14th Edition Manual, later this year.

http://www.FerrellEngineering.com

 

[DHKpeWI]

connectegr:

I am not suggesting that the tabulated values (pages 10-105 ff.) be used. My point is that the effects of the horizontal force from the axial load be combined with the vertical shear, and the bolts, plates, and welds be evaluated with the combined loads.

My reference to page 10-101 was to point out that you need not consider eccentricity on the bolts due to the vertical shear if the conditions on pages 10-101 & 10-102 are met.

 

[iponom]

I also consider local web/flange bending on the supporting member due to axial loading (at the shear plate weld)... Which most people do not and I get really interesting results..

 

[connectegr]

If welded only to a girder or column web, then I agree that a yield-line analysis is necessary to check the web. However, if the plate is welded to the girder flanges or column stiffeners, the axial load can be accounted for in the strong axis of the flanges or stiffener plates.

If the plate is welded to the column flange, relatively close to the centerline, then I don't think flange bending will control.

http://www.FerrellEngineering.com

 

[VanEng]

I am currently debating a similar situation with a steel fabricator's connection engineer. We have a large pass-through axial force adjacent to a brace bay combined with the shear in the beam. The resulting connection has 4 columns and 4 rows of bolts.

I am trying to check the extended shear tab for combined moment and axial load. What would you calculate to be the moment in the plate?

The connection engineer is suggesting that there is no moment and that all of the eccentricity is being taken by the bolts, and as such bending in the shear tab is not an issue.

Even if the shear reaction is at the end of the shear tab welded to the HSS column, we would have some bending in the plate as we go towards the bolts...

 

[connectegr]

Definitely bending in the plate and eccentricity in the bolts. Unless the beam flanges are restrained (moment connection). Eccentricity in the bolts from the weld line to the centroid of the bolt group. Gross bending to the same eccentricity. Net bending to the first line of bolts.

Also sounds like a lot of force for a HSS face welded shear plate. Did they check the column wall for concentrated force. Would a through plate not represent the load path better?

http://www.FerrellEngineering.com

 

[VanEng]

Thanks connectegr.

You are correct we do have through plates for the vast majority of these connections. The only exception is where we have pass-through forces crossing each other. Do you have a recommended detail for this situation?

We have concrete filled columns so compression is not an issue. I believe that this would help prevent local yielding of the column face under tension as well.

 

[connectegr]

How can you have a "transfer force" or "pass-through" force at a corner column? Where is it going? Is the column taking the large moment? If this is a drag force then the corner would have zero axial force at the corner column connection.

I don't know of any AISC examples for concrete reinforced columns. Chapter K provides the formula for checking the HSS wall for concentrated force.

I have designed connections for large axial load where column lines intersected. Cruciform through plates were used, with one plate continuous and the intersecting plates welded in the middle. This is not difficult at the top of column or baseplate. But, requires the columns to cut and spliced at intermediate floors.

Depending on the bracing forces, it can also be difficult to design the bracing gusset connections without through plates. The eccentricity for the vertical component is to the centerline of the column. Unlike a simple shear plate connection that considers eccentricity to the face of the column. The gusset plate is shop welded to the beam flange and field bolted to extended plate connection considering the eccentricity (or axial/moment with UFM)


The joy of connection design with HSS sections...

http://www.FerrellEngineering.com

 

[nutte]

Eccentricity in the bolts from the weld line to the centroid of the bolt group. Gross bending to the same eccentricity. Net bending to the first line of bolts.

I agree with the bolt eccentricity, and the net bending eccentricity. But I would put the gross bending eccentricity on the first line of bolts as well. By the time you get to the second line of bolts, you've removed load from the plate.

 

[connectegr]

That is the theory for net bending. But, for gross bending your are looking and the full load/moment applied to the gross area. Similar to a point load applied at the centroid of the bolt group.

http://www.FerrellEngineering.com

 

[abusementpark]

That weak-axis bending is going to result in a very thick plate.

For the OP's case of combined vertical shear and axial load, where is the weak-axis bending in the shear plate coming from?

Is it due to the slight axial load eccentricity between the shear plate and beam web centerline?