Design of baseplates with gussets

[IngDod]

Greetings, I have a question as to how to adequately design baseplates with gusset plates.. Specifically for braces. I attached a PDF showing my situation.

In general (not for baseplates) I have used the Uniform Force Method but I know this method is not applicable to situations like this. I am confused as to how the shear, and tension/compression at the gusset to baseplate interface should be obtained; and also if there will be any moment due to the geometry (Even though the axis of the brace and the column are concentric). Basically my question boils down to this.. How do I calculate the shear, axial and moment at the interfaces between gusset and baseplate and gusset and column face.

Thanks.

 

[KootK]

I believe that you can use UFM for this. It's just not the typical case that they deal with in the literature.

Assuming an HSS column, I typically assume that the work point is at the junction of the HSS wall and the top of the base plate. Then, I assume that the vertical component of the brace force is dealt with entirely in shear at the gusset / column interface and the horizontal brace for is dealt with entirely in shear at the gusset / base plate interface. Both of the shears mentioned intersect at a common point with the applied load without moments at the interfaces and, thus, UFM would be satisfied.

Using this method, there would be moment on the column. I feel that it's prudent, however, because neither the HSS wall nor the base plate is particularly stiff for transverse load. Much depends on the anchor bolt layout however.

To assume that the column takes no moment, I've used the following, more expensive, details in the past:

1) Slot the gusset plate through the HSS so that engages both sides of the column:
1) Use a WT for the gusset to reduce the flexibility of the HSS wall.

It's important to recognize that we usually don't "know" the moment and shear at the interfaces of our steel connections. We assume a path, design for those loads, and try to ensure that alternate load paths are ductile enough to get out of the way. Some assumptions are obviously better than others and there are cases where the ductility argument is a bit dubious (buckling modes, anchor bolts, etc). That's how it works though.

Your gusset plate's a little funny looking. I assume it's just rough at this point. Let us know if you need help with that.




The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[IngDod]

@KootK:
Thanks, I have tried to use the UFM for this situation but I´ve never had complete confidence in the result. Usually what I do is set the height of the "beam" to be the same as the base plate thickness. With the resulting geometry most of the vertical reaction ends up in the column and horizontal force ends up being twice in the baseplate that in the column. Since this is the UFM I assume that there's no moment in the interfaces. And yes you are right the drawing was just some quick sketch I put together so I could upload here. With my current configuration I am getting a gusset plate that is twice as Tall as it is wide, but that probably has as much to do with the angle of the brace as well as the base-plate.

For the HSS wall problem what I do is weld a reinforcing plate flat to the wall... I believe it should be cheaper than using slotted plates.. but the plates get rather thick 1/2" up to 1".

I watched one of the AISC seminars on base-plate design and the presenter was very clear in saying the UFM was not conceived to be used for baseplates... The argument was that the UFM method assumes the beam and column to be very rigid and that base-plates don't fulfill this requirement.

I still would very much like to hear inputs on this.

 

[structSU10]

here is a webinar from AISC that can be viewed for free. It touches upon UFM for base plates, and some design issues.

http://www.aisc.org/content.aspx?id=19858

 

[IngDod]

@structSU10:
Thanks, That is precisely the webminar i referred in my post. Its very good but it just points out all the issues to take account when designing.. not to actually design a base-plate.

 

[KootK]

I just watched the video. I retract my opinion that UFM is suitable for base plate design. That being said, given that the only accurate design method seems to be FEM, UFM seems as good as any of the other practical, non-correct, under capacity design methods available.

I wonder if the lecturer modeled any of those L-shaped base plates that you find at corner condition sometimes. I'd love to see the over-stress on those...

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[IngDod]

@KootK:
Yes the webminar is very good at explaining the issues with baseplates... but it also doesn't solve the fundamental problem that is how to actually design baseplates. I found it to be disappointing in that regard.

 

[SAIL3]

I usually don't fool around with trying to make the col face of HSS members handle the loads unless there is unlimited room in the budget or schedule. Also, I try and have the centerlines of members on the project be concentric. I add a reinforcing plate to the col face to transfer any loads @ that interface to carry loads to the corners of the HSS and do not count any contribution from the col face.Once I get the loads into the col sidewalls I am home free. Even if one ends up with a 1" thk pl...steel is cheap.If the brace is also a HSS memb, then I try and end up with a brace the same width of the col and add gusset pls to the sidewalls.
Design of Welded Structures by Blodgett is a must in any engrs library..it is clear, direct, practical and dirt cheap...available fron Lincoln Electric.

 

[IngDod]

@Sail3:
that is my line of thinking with hss as well... do anything possible to take the load from the column face and transfer it to the sidewalls.. I am curious however, does matching brace and column and adding gusset plates still gives you enough ductility in the connection? Im talking about seismic design mostly.. perhaps you dont have that issue in your area. It is my understanding that part of the benefit of the slotted plate connection is that its very ductile and ensures that the failure occurs at the brace rather than the connection, basically the brace buckles and the plate bends sideways to accomodate the bending... a double gusset plate would restraint this movement considerably I believe.

Also, I would like to ask you... how do you usually approach the design of the plates themselves? I'm talking about sizing the plate (width and height, not thickness...)and the load transfer between brace and column and brace and base-plate. Do you use UFM or some other method?

 

[IngDod]

Sorry for posting again but I dont see a edit button.

@Sail3: I have mister Blodgett book and it is wonderful.

 

[SAIL3]

Good point about the seismic requirement re side flexibility of the gusset pl. Right now I do not know enough about the intent or background of this requirement and would have to do some research to make a sound engingineering judgement on it. If this connection is on a project that requires seismic detailing per AISC 341, my instinct tells me to use a slot on both col faces for the gusset pl as Koot mentioned.
Otherwise, when using the reinforcing pl on the col face I would approach it as follows:
1. Make a section cut of the gusset pl @ the col face.
2. Convert brace load into shear and normal forces on this section.Normal force will also result in a
moment on the section. Design gusset for these loads.
3. To obtain load on reinforcing pl, use max normal stress on above section and multply it by gusset
thickeness to get line load on reinf pl.
4. Take a 1" strip of reinforcing pl and assume it is pinned at each end and subjected to this max
line load @ center. After it is designed check for combined max stress when combined with shear.

Two options for the width of this reinf pl.
a. Stop pl just short of the beginning of the radius on col so one can get a fillet weld in there.
b. Extend pl past beginning of radius on col so that one can get a partial flare-bevel weld and back-up
fillet weld.
The governing height would be either that reguired by the gusset designed above or as often happens by the particular geometry involved.

It is much more beneficial for an engingineer to understand the basics/background of conn design than to rely entirely on a canned method found in the codes.Blodgett is a good place to start.