Cold Form Floor Joist Connection Design Guidance

[zurch1818]

I'm trying to design a knee brace connection to this cold form joist that I can't really find any relevant information on. The section is two cold-formed channels measured to be 8" deep, spanning 13', and they are spaced 4' on center. The plans call out 60 SJ 16 with a design snow load of 23.1psf. The stud walls below them were designed for 330.2psf. With a 13' trib. width (6.5' going each way), the design load could be roughly 50psf. However, I don't really know if this is ASD or LRFD. See below for the picture and plans:

View attachment IMG_1504.JPG

From my understanding of this section, this is a 16" GA, 6" deep channel (which doesn't match the field measurement). Since it is cold-formed steel, I feel this structure was economized and putting in a knee brace connection spanning to two of these 4' on center double C members is really not the best idea. Even if it was, I don't know what the best way to attach steel angle/HSS to a cold form channel. I'm leaning more towards putting in a second member spanning between the walls and not relying on the cold form floor joists. The only information I've found on this section is from an old USG manual...but it has to do with tall wall studs and not being used as floor joists.



Can anyone provide any design guidance for my situation or at least know where I can find a load table to get a feel for how these floor joists are performing in the original condition? Thanks for your help,

 

[XR250]

I think posting a section may be helpful

 

[zurch1818]

I think posting a section may be helpful

I'm not sure I follow. Here is what our drafter preliminarily came up with for this site:



This mount would be attaching in the bays of the 6" light gage studs spaced at 16" O.C. The kicker would come up to member that spans between the two Double C roof joists marked as "60 SJ 16" (discussed above) that are spaced at 4' O.C. We can go back into the field to try to get a measurement on the C channels but I was hoping for some sort of load table for this application but I wasn't able to find it.

I'm anticipating there being a significant amount of ice load on this mount, so if the joist was at 100% in the original condition, this type of connection may not be the best (as snow and ice are in the same load combination). We do have a precast floor below, so a lot of the load can go down there...but I need to have some sort of connection up top to help take out some of the lateral load as I was having issues with edge distances and this mount acting like a huge cantilever.

Maybe it would be better to have a strong 2x6 span in the web of this stud and reinforce the joist? I'm guessing that I should be able to to find a section to span this distance to argue we aren't weakening the original stud? I'm hoping to get some advice from someone that deals with light gage.

 

[XR250]

I don't think your are presenting your situation in a way that people can understand it. (at least I can't)

 

[KootK]

so if the joist was at 100% in the original condition, this type of connection may not be the best (as snow and ice are in the same load combination).

It might be feasible. Snow will be pushing the joist downwards and ice on your mount will be pushing it upwards. So the combination won't be entirely additive as far as the joist stresses are concerned. The mount loads may cause the bottom flange of the joists to go into compression. If that is the case, they may need to be braced against lateral torsional buckling. Conceivably, your cross member (HSS?) might be used for this purpose.

I would study the connection between the joists and the wall and see if you can make that work without your kicker. Even if you add the kicker, I'm not sure that you'll be able to entirely shield that connection from load. A shallower brace angle would feel a bit better if that would be viable.

As far as load tables go, you'll find a pretty robust specification guide here: SSMA

 

[zurch1818]

I don't think your are presenting your situation in a way that people can understand it. (at least I can't)

Got it. I don't work in cold-form or really building design, so I'm trying my best.

It might be feasible. Snow will be pushing the joist downwards and ice on your mount will be pushing it upwards. So the combination won't be entirely additive as far as the joist stresses are concerned. The mount loads may cause the bottom flange of the joists to go into compression. If that is the case, they may need to be braced against lateral torsional buckling. Conceivably, your cross member (HSS?) might be used for this purpose.

That makes sense with the ice load being a diving board. I'm guessing putting in some bracing to help with the LTB will be helpful. I wasn't quite sure what section to really use. I primarily specify square HSS for this type of application...but with it being cold form, the standard connections I'm used to with 1/2" to 3/4" A325 bolts don't really work. I could see using an angle or something different if someone wants to recommend what would be more typical in this application.

I would study the connection between the joists and the wall and see if you can make that work without your kicker. Even if you add the kicker, I'm not sure that you'll be able to entirely shield that connection from load. A shallower brace angle would feel a bit better if that would be viable.

I'm in process of trying to figure this loading out right now in terms of the angle and how the connection does. I wanted to do some prework last week before diving into the calcs this week. We have the flexibility to move this connection to shallow up the angle. I can see that a shallower the angle, the more the joist gets put into axial and the bending/flexural component becomes less significant. I'm a little worried about the top track connection without the kicker.

As far as load tables go, you'll find a pretty robust specification guide here: SSMA

Thanks. I'll take a look at that resource. It will definitely be helpful to at least gage performance of what it could have been designed for. When briefly looking at it, I'm seeing that they have spacing at at most 24" in those tables. Since I have joists that are doubled up at 48" OC, it seems plausible to treat it as each joist have a 24" trib width. The only thing that I can think of with this is the LTB requirements not really working as they specify, but it at least gives me a feel. They are tied together, so the compression flange is sort of "braced". I'm still going to dig into it further and maybe I'll find some other goodies in there. Since the field measurement doesn't match the plans, I don't really know what section got used and I can send someone back out to get it but I will try to see if I can proceed without it. I do have the full design loads of the members, so I hope I can stay within the IEBC requirements of this member to where I don't have to do rigorous checks on it (5% for gravity, 10% for lateral).

 

[KootK]

Actually, I see now that I misread some things:

1) You've got an HSS post in the wall so you're not using the stud / track.

2) The Ice load will produce downward load on the joists after all. Botched that...

 

[KootK]

I primarily specify square HSS for this type of application...but with it being cold form, the standard connections I'm used to with 1/2" to 3/4" A325 bolts don't really work

Yeah, cold formed framing often connects better with screws.

Is welding to the CFM an option here? You might be able to do that if the material is 16ga or thicker.

Will there be conditions where you would have those HSS on both sides of the back to back channels?

 

[zurch1818]

I've got (2) square HSS members that I'm putting between the 6" wall studs @ 16". I'm going to have 2 kickers that are spaced at 4' in the bays (green). These are going to connect directly to the HSS posts that are getting embedded in the wall. These kickers will then connect to 2 members (in purple) spanning between the yellow roof floor joists that I've been referring to. I don't really know what the best material/section is for the purple members.


This is what I'm using for my frame to try to decouple the loads as best I can. I'm trying to build in as much redundancy as possible with this design because I ultimately don't know how the connection to the building will match my FEM model. Once I add more fixity/rigidity up top to get the deflected shape in order and fix the edge distance issues in the precast hollowcore slab, then I start getting more force up top. It's a structural FEM whack-a-mole game. It's a game I play all the time, so I'm not too worried about the hot rolled steel.

Yeah, cold formed framing often connects better with screws.

Is welding to the CFM an option here? You might be able to do that if the material is 16ga or thicker.

Will there be conditions where you would have those HSS on both sides of the back to back channels?

Hopefully my crude diagram gives a better picture of what is going on. We should be able to add as many purple members as we need to. We should also be able to stagger them. Welding to the CFM is definitely an option here too. I've never specified it for CFM before, but seeing the plans call out 16GA, it should in theory be possible. In the diagram above, this would be where purple meets yellow. Do you have any resources for me to research welding to cold-form? I don't feel I will need that much weld as the weld is probably going to be much stronger than the CFM.

Not that it is really pertinent to this question, but in the other spot on this building, we have open web steel joists in which I'm welding on a brace to get load to the panel point, so the contractor has the ability to weld on-site.

 

[KootK]

A little whacky but something to consider. It would add no bending to the joist, only axial tension.

 

[zurch1818]

A little whacky but something to consider. It would add no bending to the joist, only axial tension.

I'm trying to wrap my head around this. That's a second HSS going vertical and attached with some sort of plate out between the wall studs and then connecting straight up to a 4' long CFM? With the member being so vertical, can't it only put bending into the roof joist (which I now realize that I've also been calling a "floor joist") and not much axial load? I could see that the overall bending being pretty minimal because it is so close to the end of the span. I'm just not sure how well the FEM model with perform with this as I'm anticipating some issues at the base because the decoupling distance is decreasing. The top is essentially becoming less stiff and stiffness attracts force.

I've been wrong more times than I can count, so we will find out tomorrow when I get back in the office (assuming the kids don't get sick again).

 

[KootK]

That's a second HSS going vertical and attached with some sort of plate out between the wall studs and then connecting straight up to a 4' long CFM?

Exactly. Think of the column as continuous from foundation to the elevation of the centerline of the joists. The arrangement is that, just with a local offset.

With the member being so vertical, can't it only put bending into the roof joist (which I now realize that I've also been calling a "floor joist") and not much axial load?

This should be the arrangement that would do the most to minimize the additional bending in the roof joist. That bending will be the lateral reaction at the top of the post multiplied by the distance between the clip angle connection fasteners and the roof deck. Pretty small.

I'm just not sure how well the FEM model with perform with this as I'm anticipating some issues at the base because the decoupling distance is decreasing.

My arrangement may increase the column base shear a bit. That increase should be small and paid for in spades by the improvement at the top of the system where things will be much more critical. The strategy is to minimize the impact to the delicate bits at the expense of increasing demand on the robust bits. At the end of the day, all of the load has to go somewhere.

I would encourage you to not let yourself get lost in the FEM. In my mind, this is a situation where you want to design and control the load path rather than "discover" the load path via FEM. Were I doing the modelling, I probably wouldn't even bother to model the column offset that I propose. I'd just treat it as a continuous, pin ended posts as I described above.