Slip-Critical Connections in Seismic Regions 1

[DCBII]

AISC 341 requires connections in the seismic force resisting system (SFRS) to be detailed as slip-critical if R>3. I've traditionally thought of a seismic force resisting system as a SMF, SCBF, BRBF, etc, including anything like collectors directly attached to the system. I've always used slip-critical connections in these elements since they're likely to undergo large inelastic deformations as the system yields.

I've never really thought of diaphragms as requiring slip-critical connections, since they are supposed to remain elastic, and most of the deformation is occurring elsewhere.

I've designed a special reinforced masonry shear wall building with a flexible diaphragm steel roof. The beams in the roof act as the diaphragm's continuous cross-tie. There is a girder running down the center of the building that the beams frame into. Do the beam connections into/across the girder need to be slip critical? Is that the intent of AISC 341? Or is that over-specifying slip-critical connections? Slip-critical connections cost about 3x as much, and they would be otherwise unnecessary on this project. I suppose the beams are technically part of the SFRS, but I'm not sure AISC intended this provision to extend to the continuous cross-ties. To me the primary seismic force resisting system is the masonry shear walls.

Thoughts?

 

[DAVE9999]

AISC 341 does not require slip critical connections, the bolts can be designed as bearing type connections but the faying surfaces must meet the requirements for a class A slip critical connection and the bolts must be pretensioned (see 341-05 C7.2 and 341-10 D2.4).

 

[DCBII]

You're correct Dave9999. Thanks for the clarification. They are "psuedo-slip-critical" connections since the hole isn't slotted or oversized, and the calculations have been run for bearing. But from the detailing standpoint, by the time you prepare a class A surface and add pretension to the bolts, you've basically got a slip-critical connection with all its associated cost impacts.

The question I'm really asking doesn't relate to the specifics of design/detailing. Rather, it relates to whether the intent of AISC is to apply this requirement outside of vertical frames and their components and collectors. Does it apply to my diaphragm cross-ties?

 

[jittles]

I can't say I have a rational explanation or a codified basis, but I've never seen diaphragm cross ties included under SFRS bolting requirements. I've worked on many a steel building and nearly all of them have been in high seismic regions.

 

[JoshPlumSE]

I think the intent is really fully pre-tensioned bolts. If you've got a part of your lateral force resisting system for seismic, you want it to be properly tightened, not "snug tight".

I've always interpreted this as any part of the lateral force resisting system. We used a lot of horizontal bracing in the heavy industrial design I used to do (because we didn't have diaphragms). Every one of those was pre-tensioned. Heck, in that work, I believe we even pretensioned our gravity only beam connections. But, that may have just be a company preference rather than a true code requirement.

 

[DCBII]

My background is also in heavy industrial structures. Most heavy industrial structures I've designed were done based on ASCE 7 Chapter 15, with an R value of 1 to 1.5. There the provision definitely wouldn't apply. I did see contractors using twist-off bolts from time to time, but we only required snug tight. JoshPlum, did you have them do any kind of preparation to the faying surfaces?

 

[skeletron]

@JoshPlum: I believe Clause 22.2.2.2 in S16-14 addresses when pretensioning is required (LINK)

 

[DAVE9999]

"you've basically got a slip-critical connection with all its associated cost impacts", this is not accurate. I'm not trying to be a pain but a bearing connection has about 40% more capacity than a slip critical one in shear. You will use a lot more bolts with slip critical, hence the added cost.

 

[TLHS]

DCBII, what are you building where your heavy industrial structures have R values of 1 or 1.5? That's, like, unreinforced masonry, unusually tall framed structures, or ordinary moment frames in the absolute worst SDC.

I've done heavy industrial structures with really low R values to meet performance requirements once or twice, but that's the exception rather than the rule.

 

[JoshPlumSE]

We generally used OMF and OCBF. So, R was greater than 3. I'm sure the surface preparation and painting was covered in our general notes and specs. But, I don't recall how that was called out.

 

[DCBII]

TLHS, for many industrial structures using the R factors out of chapter 12 of ASCE 7 is overkill, and not feasible. For example, there aren't any seismic systems in AISC 341 that meet the needs of mill buildings, which utilize truss portal frames for a lateral system. There are special truss moment frames, but last time I checked the span limit for those was something like 65 feet. Even if you designed such structures to be inelastic, what might happen to the bridge crane if the drift got excessive due to inelastic column rotations? There's a myriad of other reasons for low R values, but if you want more info I'd refer you to the attached article. AISC has published one or two similar articles in Steelwise in recent years.

 

[TLHS]

Fair enough!

Large, gravity efficient buildings I would understand. You start hitting span and height limits and sometimes have awkward load paths where it's hard to define the LFRS.

To me the term 'industrial structures' encompass a lot more than that so I was a little confused. You can often either reasonably assign a structure to an existing category, or rationally find an analogue and make sure you detail for any weird situations. I don't mean R values of 6 or anything like that, but R values of 2 or 3 are pretty doable in a lot of cases.