Cantilever beam on shear connection. Maintenance insisting it works cos they use it 2

[delagina]

I have a 4' cantilever beam connecting to an existing beam. Maintenance said to use a detail connection that they've used many times in the past. It's a shear connection. How do I go about this? Hard to argue with something that is working for many years.

 

[delagina]

Load is minimal so the "moment capacity" of the shear connection will probably work. I'm conflicted with this thought. When do you consider the moment capacity of a shear connection.

 

[WARose]

How exactly would maintenance know what the connection is good for? That's the argument I would make. A 4' cantilever with a end shear connection could never be defended if something goes wrong.

I would think the deformation and girder rotation would blow it out of the water anyway.

 

[delagina]

His argument is they've use it all over the plant.

Shear connection has some moment capacity that's why it's working. Load is minimal.

 

[WARose]

If you can make it work.....it's your choice to use it. I'd check it stem to stern though. (Starting with the web of the girder it is connecting to. Webs don't have a lot of out-of-plane capacity.)

 

[delagina]

I guess my mindset always been. It's shear connection, it got zero moment capacity. Even all our connection spreadsheet are like that.

 

[TehMightyEngineer]

What kind of shear connection is it?

Ian Riley, PE, SE
Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[rb1957]

1) what do "maintenance" know about structural analysis and code requirements ? Their options should be read with this in mind.

2) Just because something hasn't failed (yet) doesn't mean that it meets the requirements.

3) Do your analysis, show it good (or not).


another day in paradise, or is paradise one day closer ?

 

[KootK]

We're talking about a moment connection to the side of the girder rather than the end, right?

When do you consider the moment capacity of a shear connection.

When you have to. What are we talking, shear tab, single angle, double angle? As you say, most shear connections do have some moment capacity so if the loads are small enough, it's doable. Watch your LTB bracing at the connection though.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[delagina]

Clip angle to an existing beam. Where can find moment capacity of this case? Honestly a part of me agrees with him. A part of me always assume shear connection as zero moment capacity because that's what we always assume in analysis.

 

[dik]

When do you consider the moment capacity of a shear connection.

For Cantilevers... do you have a sketch and loads?

Dik

 

[WARose]

By the way, if this thing is so lightly loaded......why not just a piece of Unistrut? I would think that would be cheaper. (And more stable. You could run it across the top/bottom flange of the girder.)

 

[rb1957]

the situation is described as a cantilever. How can the end of a cantilever be considered as a shear joint ?

How do we know the loads are small ? (other than the description of a 4' span ?)

The torsion effect on the supporting may need to be considered too.

another day in paradise, or is paradise one day closer ?

 

[CANPRO]

I was thinking the same thing as rb1957 regarding the torsion...you probably have some nominal moment capacity from the shear connection, but if you get that combined with incidental torsion or lateral loads, you might be in trouble.

Is it a single clip or double? Beam sizes and expected loads would be helpful.

On a side note - you had a thread a couple days ago about extending some purlins. Lots of people chipped in with good suggestions but you stopped responding. Same as this thread, you don't give any information that would lead to a meaningful discussion. If you're looking for help and people are willing to help, its only right you supply proper info and answer the questions.

 

[JoshPlumSE]

rb1957 -

In reality there is no such thing as a perfectly "pinned" or perfectly "fixed" connection. These are just convenient terms we use to idealize our connections. In reality all connections fall somewhere in between the two idealized cases.

All the OP is trying to say is that he's got a situation where a connection that he would have normally designed as an "idealized" pinned connection is required to take some moment. And, he's looking for help in how to analyze it. Now, in order to get a more detailed response he should probably show a sketch of the connection.

I would offer the following advice....following an assume load path (in reverse).
1) Can the girder web resist the forces/ moment imposed from the clip angle? I would look at "plate elements subject to out-of-plane loads" from AISC's 15th edition steel manual.
2) Connectors between the clip angle and the web of the girder. Welds or bolts. Should be straight forward.
3) Can the clip angle itself resist those moments? Usually only an issue if the outstanding leg is very thin or long. There are AISC formulas checking for buckling of shear tabs due to an applied shear or moment.
4) Connectors from the cantilevered member into the clip angle.
5) Serviceability. What kind of deflection is this going to produce? Assume bolts can slip in the worst possible way with their bolt holes. Maybe assume some deformation of the bolt holes. Project that rotation into a deflection at the tip of the cantilever. Add it to the traditional cantilever deflection calcs.

 

[KootK]

Is the name of the game here avoiding any welding? If a shear tab welded between girder flanges and then bolted to the outrigger element is an option, that's pretty easy. If you'll have to evaluate a bolted-bolted, partial height, single angle connection, that'll be tough. I imagine it's a prying action / yield line analysis for strength. Not sure you'll ever get a good handle on joint stiffness unless WARose is willing run an FEM model for you.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[WARose]

..... unless WARose is willing run an FEM model for you.

Always willing (as long as it's not too time consuming ).

 

[MrHershey]

If it's real light load you may be able to justify it as a moment connection by resolving the moment into horizontal shears in the bolts and checking that against bolt strength. If you can't justify it that way, you probably need to put your foot down.

And very important to understand that 'works in real life' is not the same as 'works on paper'. In my interactions with lay people and people with different roles in construction profession, I've found there's a big disconnect in understanding there. And it's difficult to argue that things that have been used successfully for decades don't work because they clearly do for what they've seen thus far.

But it's important to remember that we don't design structures just for things that happen only every few decades. We design for the absolute worst day of a structure's life. Things that don't pencil out often end up working in real life because the vast majority of structures have not seen the load they were designed for yet. Our live loads are significantly larger than you'd get if you actual measured typical uses. Our return period for strength design of wind in normal buildings is 700 years. Our return period for maximum considered earthquake is 2475 years. Even our dead loads which we should know with near absolute certainty get inflated. And that's just on the loading side, that's not even getting into material and design safety factors.

 

[rb1957]

I know there's nothing in nature that is perfectly fixed, but it is possibly the typical analysis assumption.

However, I can't see a cantilever with a shear joint.

In a perfect world I guess we'd calculate a partially fixed cantilever (depending on the torsion stiffness of the beam) but I'd use a fully fixed cantilever as a conservative assumption. In any case the moment at the the end of the cantilever is predetermined (so as not to say "fixed") by the applied load; the degree of fixation only affects the amount of rotation that occurs there and, of course, the displaced shape of the the cantilever.

another day in paradise, or is paradise one day closer ?

 

[jike]

Another possibility is that it is working but has an inadequate factor of safety legally mandated by the building code.

I can see it possibility working with a shear connection if the load is very, very low but if it doesn't check out explain to them that you (and the owner) have a legal responsibility to provide an adequate safety of factor.