Joist Staggering 2

[EngStuff]

This isn't project specific, I looked throughout the site and all over, I only found one thread which someone suggested to stagger joists on a particular project.

My question is more of besides constructability and maybe the CG being annoyed about it. Assuming we have a roof with a steel girder which has steel joists on the right of it spaced 6'-0 on center. Joists on the left spaced 6'-0 o.c. but the right and left joists are staggered from each other about 3 feet away. If torsion on the beam was not a concern or taken into account. Would we have any other issues we have to deal with? As in diaphragm design or force transfers. I myself don't see an issue, but doesn't hurt to see what everyone else thinks about this? Perhaps I might not see something that could be an issue.

 

[KootK]

1) I think that it's more constructable to align the joists either side of the beam to make it easier for the guy doing the deck fastening. I'll go out of my way to ensure that all of my beam widths are wide enough to accommodate this.

2) I've seen a lot of typical detailing from eastern Canada (I'm in the west) indicating that they do often stagger their joists. They do it 6"-8" though, not 1/2 a joist spacing.

3) I don't see that the stagger compromises diaphragm behavior so long as you're designated chords & collectors are continuous where they need to be.

 

[XR250]

KootK,

What is typically the minimum beam width to achieve alignment?

 

[jayrod12]

I've heard a minimum of 140mm or 5.5". That's why a W14x33 and W16x39 are so popular.

 

[XR250]

That's why a W14x33 and W16x39 are so popular.

Can't be too popular as I have never heard of those sizes

 

[KootK]

Yup, 5.5" for me too.

 

[jayrod12]

Sorry, mixed up my metric and imperial

W14x22 and W16x26? Those should make more sense

 

[XR250]

5.5" - good to know. Thanks

 

[CivilSigma]

What is meant by staggering? Do you mean overlapping? Thank you.

 

[jayrod12]

They mean that the joists on either side of the beam are not inline. They're staggered from one side of the beam to the other. Perhaps this would mean a small overlap when the beam flange isn't wide enough to accommodate the two joist seats, however I feel that's a rare case where a steel beam is less than 5 1/2" wide supporting steel joists.

 

[CANPRO]

I've seen a lot of typical detailing from eastern Canada (I'm in the west) indicating that they do often stagger their joists. They do it 6"-8" though, not 1/2 a joist spacing

Eastern Canada perspective here - for regular bays the joists would almost always align, but I have seen the small stagger you mention on occasion.

The 5.5" minimum width makes sense as a practical minimum if you add up all the bolt edge distances you need to maintain. If you want a min. 1.25" edge distance for your bolt, you need to maintain that on each side of the beam and with each joist shoe. With a nominal 1/2" erection gap between joist shoes you're left with min width = 4*1.25" + 0.5" = 5.5"

Sorry, mixed up my metric and imperial

W14x22 and W16x26? Those should make more sense

The plight of a Canadian engineer - I feel your pain. My brain is a painful mix of metric, imperial, and all of the fun conversions that go with it. Those two beam sizes are so popular I might actually be able to recite their section properties without looking at the book.

 

[BAretired]

The 5.5" minimum width makes sense as a practical minimum if you add up all the bolt edge distances you need to maintain. If you want a min. 1.25" edge distance for your bolt, you need to maintain that on each side of the beam and with each joist shoe. With a nominal 1/2" erection gap between joist shoes you're left with min width = 4*1.25" + 0.5" = 5.5"

I'm curious about your mentioning bolts. I have always seen joists connected to steel beams by welding, but I would prefer bolts because they are not so likely to be omitted as welds, but I expect bolting adds to the cost.

Photographs of the Station Square collapse in Burnaby, B. C. suggested to me that some of the joist-to-beam connections in the failed beam had not been welded at all. This was not mentioned in the report of the collapse but I believe it to be true. I think that bolts would not be as easy to miss, either by the installers or the inspectors.

BA

 

[jayrod12]

Actually BA now with the advent and increased use of machinery to fabricate the steel, bolted joists are quite prevalent. At least where I practice in central Canada. So we give them the option of doing either. However it does require that we hold our beams slightly less than 100% to ensure that we have capacity leftover after the section loss for the bolt holes. Where the beams are in positive bending, it's less of a concern because the bolts provide a path for the compression stresses. Where in negative bending there's around a 10-15% reduction is section due to the bolt holes, so we design appropriately.

 

[CANPRO]

BA, my preference is bolted connections because the joist can be immediately secured once placed and released from the crane.

I don't have the reference on hand right now, but I believe once the joist spans further than 60' there is an OHSA requirement that the joists must be bolted.

 

[KootK]

Where the beams are in positive bending, it's less of a concern because the bolts provide a path for the compression stresses.

That sounds dubious to me. It would take a fair bit of strain to close a 1/16" over sized bolt hole around a bolt to engage the surrounding flange. By the time that you'd reached that level of strain, I suspect that the beam would have reached it's usable capacity in some fashion.

Where in negative bending there's around a 10-15% reduction is section due to the bolt holes, so we design appropriately.

I've typically found the reduction to be less, sometimes nothing at all. Are you taking advantage of the ability to mobilize Fu locally rather than Fy in your calculation?

 

[XR250]

Seems like the fabricator and contractor would have to be super accurate to get all these bolt holes placed accurately - especially if they were bolted at each end with the less than perfectly plumb walls. Welding looks a lot more forgiving.

 

[jayrod12]

That sounds dubious to me. It would take a fair bit of strain to close a 1/16" over sized bolt hole around a bolt to engage the surrounding flange. By the time that you'd reached that level of strain, I suspect that the beam would have reached it's usable capacity in some fashion.

Honestly, this was an inter office memo thing that was done by one of our senior designers. You're likely correct, although at the same time, when it comes to ultimate strength I'd bet it still suffices for a load path.

I've typically found the reduction to be less, sometimes nothing at all. Are you taking advantage of the ability to mobilize Fu locally rather than Fy in your calculation?

Same as above, the guidance came from a senior designer years ago. It's essentially an office "rule". I doubt they did look at it in those terms but rather just looked at net area reduction. I'll try to find the memo and see if I can gleam any further information from this.

 

[CANPRO]

Seems like the fabricator and contractor would have to be super accurate to get all these bolt holes placed accurately - especially if they were bolted at each end with the less than perfectly plumb walls. Welding looks a lot more forgiving

This is true, but really no different than if your framing was all beams with bolted connections. You might have to slow down the erection from time to time to weld a mis-aligned connection, but the economy of all bolted connections far out weighs the occasional welded connection. The name of the game in the steel business is accuracy. As the saying goes in my line of work "if you're off an inch, you're off a mile".

And if you're framing your joist into a wall then it is likely a welded to an embedded plate, so you do have some flexibility on one end.

 

[KootK]

Seems like the fabricator and contractor would have to be super accurate to get all these bolt holes placed accurately - especially if they were bolted at each end with the less than perfectly plumb walls. Welding looks a lot more forgiving.

I've always hoped that welds would help rectify some of the beam torsion that I rarely bother to check. Maybe bolts would be good enough in that respect too...

You're likely correct, although at the same time, when it comes to ultimate strength I'd bet it still suffices for a load path.

There's a path for the flange stresses, yes. The question, from my perspective, is whether or not you can realistically hope to mobilize that path before other terrible things happen. At the level of flange strain that it would take to compress a 13/16" length of flange by 1/16" is substantial. I worry that our normal b/t local buckling limits would not be applicable at that level of strain. Do our b/t ratios assume a mild excursion into plasticity or all the way to strain hardening as would be the case here at 5X+ the level of strain that initiates hardening? I really don't know.

 

[XR250]

You might have to slow down the erection from time to time to weld a mis-aligned connection, but the economy of all bolted connections far out weighs the occasional welded connection.

Does it? Both joists and beam have to be drilled - obviously in the shop, but still. Then the guys in the field have to line things up and install the bolts. You already have a welder on site, it don't take long to put done a few inches of weld on each side. I imagine it depends on the scale of the job, number of bays etc.