Racking Connection to Roof Joist for Lateral Support 1

Honestly, I'm not sure what I'm looking at, what the load path is, and why the racking needs to be braced. The plan doesn't match the elevation and it looks like the wall is already braced.

Based on how much I understand this, I think your idea of L2x2 is fine. They can be cut to fit and welded. I wouldn't use cables because they won't work in compression, and the geometry here makes it so that compression is needed. I wouldn't use a turnbuckle and rod system for the same reason, although it could work, but it wouldn't be any easier to connect than L2x2.

Personally, I'd weld a continuous L3x3 or something to the bottom of the joist, kind of like bridging, and find a way to connect from there. At least the geometry will be simplified to only the direction you need in bracing. The 3D problem becomes a 2D problem and the angles will be easier to connect.

@Milkshakelake

Thanks for the response. I admit I did not provide adequate background information. The existing racking is being modified such that two of the bays from the existing racking system will be removed to allow for space required for other use by the building owner.

Based on my analysis, I found that even the existing racking setup (6 x 6 grid with roughly 5 ft spacing in each direction) is only braced in one direction but not in the other. I am trying to brace the framing in the direction that currently does not have "X" or "Diagonal" bracing by connecting the two corners of the framing to the roof. The load path for lateral movement in this direction will be through the member into the roof, roof deck, and global lateral resisting system. The reason I am thinking of attaching the racking to the roof joist is that there is nowhere I can add vertical "x" or "diagonal" bracing in that particular direction as it will block the way for operators to get in/get out to place/remove pellets.


You had suggested welding a continuous L3x3 to the bottom chord-like bridging. I was thinking of achieving this by using an angle as shown in the sketch below. I am just concerned about the torsion that I will end up introducing in the channel due to the connection offset. The amount of torsion is likely going to be relatively small say 3-5kNm/m. I am thinking I can reinforce the joist by adding vertical members at the panel points.

Thoughts?

Yes, you got it. That's kind of like what I was thinking. That will work. If you're worried about torsion, the torsion will be shared by several joists with that channel, so it's not really a concern.

I was thinking about this over the weekend. What is the loading on the post? If there is considerable load, it might be worth it to add some x-bracing for torsion.

I think that your goals here ought to be:

1) Engage the joists laterally along their own axes.

2) Try not to engage the joists vertically. This, to avoid:

a) The racks holding up the roof and / or;

b) Roof load causing your racks to tilt over if diagonal braces are used without a movement joint.

3) I've proposed what I feel is a simple solution below:

a) Extend the vertical posts up through the truss bottom chords.

b) Capture the posts laterally on both sides with angles welded to the trusses but not connected to the posts.

c) Maybe add some truss bridging near the connection for good measure.

I like KootK's solution. There is still the torsion problem, so I'd definitely use his last point about adding some truss bridging. Also to add to his idea, if you use vertical long slotted holes to the angles, you will not engage the posts for vertical loads.

MSL said:

There is still the torsion problem, so I'd definitely use his last point about adding some truss bridging.

Click to expand...

I don't actually see a torsion problem with this. Or, at the least, not the torsion problem that OP mentioned which was a result of the vertical offset between the point of load delivery from the rack and the back of the channel. With my detail, I see the horizontal load getting transferred primarily at the horizontal legs of the angles, just below the bottom of the bottom chord. With respect to that eccentricity, I suspect that the angle connections to the post and joists would rectify that handily.

My intent for the bridging was really:

1) Rectify any incidental, out of plane lateral load on the joists that may come about.

2) Brace the tension chord of the truss where we might be introducing some local compression into it. I'm not sure this is even a legit demand however.

3) Just kind of feels like a thoughtful thing to do.

If you load the bottom chord laterally and not the top chord, isn't that torsion? The depth of the joist is d, the incidental lateral force is F, and the torsion is dF. It might be a small load, but OP didn't specify, so I'll assume it's a concern. You also mentioned an incidental load, so that's what I mean. That would be a much bigger concern than local compression.

Anyway, I think we're on the same page, just discussing semantics here. The end result is that your detail is gangsta good.

gangsta good... I'll take it.

Primary load = parallel to joists = no joist twist.

Incidental load = perpendicular to joists = a bit of twist.

Check.

Thanks to both of you guys.

I really like KootK's suggestion. But unfortunately, I had to think up something different as the owner did not allow attaching anything to the roof.

My end solution was to add additional framing members at the top of the racking connecting the posts and provide diagonal bracing at one end. This was determined to be sufficient to control lateral displacement and keep the loads within the acceptable range.