The multiple editions of AISC's manual. 3

[StrEng007]

I've met an array of engineers that all have their favorite version of AISC's manual (most importantly the Specification). This post is not about the benefits of one versus another. Rather, I'm trying to understand how other engineers fulfill the obligation of using the latest standards adopted by code. By default, it seems like we should all have multiple editions of the manual/specification and know what changes exist in each. However, some people strictly utilize the 9th Edition ASD manual, others the 15th edition, and some in-between, etc.

From what I can tell, subsequent editions of the manual have been focused on creating the lightest and most efficient structures. There are more provisions provided today that will allow certain scenarios to be calculated with a "refined" factor of safety. If that's the case, why doesn't AISC provide exceptions that allow interpretations of older codes.

Can anyone share on how they manage the ongoing updates to each edition?

 

[271828]

That's simply not true. There are always binding contract documents that require adherence to the standardized methods.....

That's only if the EOR says to design connections per specific design guidance, right?

Otherwise, WillisV is correct IMO. If nobody tells the delegated designer to use the single plate procedure in Part 10, the delegated designer can use another method as long as the Specification is satisfied.

 

[WillisV]

@271828 - bingo.

 

[271828]

I just looked at a couple of sets of structural drawings for connections projects I've worked on. I don't see any note that would've limited me to procedures in the Manual Part 10 and so on. I have seen more restrictive notes, but couldn't find one quickly.

It's similar to base plate design. The Specification gives the available pressure in the concrete, plate bending strength, and anchor rod strength checks. For the particulars of the calculations, I can pick the elastic method in Blodgett, the DG1 method, use finite element analysis, or whatever other method I can find. I don't have to use the Manual Chapter 14 or Design Guide 1.

 

[StrEng007]

Taking a couple steps back here, I've got some questions on procedure that I think will be beneficial to discuss:

1. Who's responsibility is it to determine whether or not to utilize specific design criteria in the overall design? It clearly isn't an agreement between the EOR and the owner of the project. An owner wouldn't even know what that means.

a. If the decision is to be made by the EOR, are they free to ignore any guidance that is not in the specification?

2. You mentioned this being outlined by the EOR to the delegated design engineer. Where I've worked, I've seen many EORs handle their own connections as the owner of the project doesn't want to pay two engineering fees (1 for the EOR on the design, 2 for the sealed engineering applied to steel shop or fabrication drawings with connection details). In that case, the EOR would have to decide how to handle the design of any extended single plate connections before a detailer ever set eyes on it.

 

[JoshPlumSE]

Who's responsibility is it to determine whether or not to utilize specific design criteria in the overall design?

I'd say that this is the responsibility of the "owner's representative". That will often be the architect on a project, or a project manager employed by the owner. I've seen projects where it's the structural, mechanical or electrical engineer. It all depends on what the primary motive of the project is.

This is usually merely a requirement that it meet the code requirements of the authority having jurisdiction over the permitting process. If there are multiple methods of design that can be used (ASD vs LRFD, for example) then it is usually up to the EOR to choose the method most appropriate for the project.

Where I've worked, I've seen many EORs handle their own connections as the owner of the project doesn't want to pay two engineering fees (1 for the EOR on the design, 2 for the sealed engineering applied to steel shop or fabrication drawings with connection details).

Yes, this is a more common practice on the west coast. But, there are still plenty of cases where the steel fabricator / detailer can save themselves (and the owner) a lot of money on a project by designing the redundant gravity only connections themselves. Though this is more common out east as I understand it.

I hope that we (as a profession) are getting better at exchanging information between engineers and fabricators than we were in the past. I think this is especially true for engineers / fabricators that have worked together multiple times.

 

[dold]

Yes, this is a more common practice on the west coast. But, there are still plenty of cases where the steel fabricator / detailer can save themselves (and the owner) a lot of money on a project by designing the redundant gravity only connections themselves. Though this is more common out east as I understand it.

Can confirm, having worked in both markets.

East coast: almost all connection design is delegated, including moment connections and braced frame connections. Other "non-primary" structural elements such as stairs and metal studs are also commonly delegated.

West coast / Rockies: almost all, if not all, connections (gravity/moment/brace) are designed in house - excluding "specialty" connections such as BRB and SidePlate connections since they are better/solely equipped to do that. Stairs, metal studs, etc are also commonly included in contracts but not always. Most owners do choose to have EOR design them because it cuts out a bunch of coordination and what have you.

Delegated design is a double-edged sword. Convenient? Yeah. Always ends up with me saying "I should've just done this myself"? Also yeah.

 

[phamENG]

Delegated design is a double-edged sword. Convenient? Yeah. Always ends up with me saying "I should've just done this myself"? Also yeah.

East coaster here. Yes, yes, and yes. This is one of the reasons that I'm working to move away from delegated connection design. By the time I've taken the time to consider all of the necessary items to coordinate the connection (design loads, etc.) and make sure that the connecting elements of the columns and beams work (ever had a connection engineer come back and tell you that your HSS wall is too thin for a moment connection? It's embarrassing...), I might as well size the bolts/welds and put it in a schedule with a typical detail.

 

[StrEng007]

I'd say that this is the responsibility of the "owner's representative". That will often be the architect on a project, or a project manager employed by the owner. I've seen projects where it's the structural, mechanical or electrical engineer. It all depends on what the primary motive of the project is.

I've been on projects with an owner's representative that was not part of the licensed design professionals providing design work on the job. I've also been on projects under the direct supervision of an architect. In either case, I haven't seen either one define any sort of requirement for design besides providing the applicable building code that the design was to satisfy. In some cases, not even that was supplied as we had to determine, on our own, the applicable code at the time of permitting. In my experience, I've never come across an architect who was familiar enough with AISC (as an example) that would be able to require the torsional checks for extended single plate connections.

From my experience, a majority of projects have been completely unguided and it was up to the structural engineer to satisfy all requirements as best as they could. If you are lucky enough (or unlucky), you get comments from a building department review in the areas that were lacking.

East coast: almost all connection design is delegated, including moment connections and braced frame connections. Other "non-primary" structural elements such as stairs and metal studs are also commonly delegated.

This hasn't been my experience in the east coast. When scheduling your design loads, do you provide a required moment, shear, axial, or whatever applicable scenario for each connection? Do you specify when loads occur in the same case or if you want the capacity to satisfy ASD or LRFD? It may seem strange that I'm asking, but I haven't experienced delegated steel connection loads.

Does anyone else feel like they've spent so much time learning demands, analysis, and design for a variety of materials and types of structures, that the art of practicing engineering is somewhat of a convoluted mess when it comes to jurisdictions, project politics, and economical decisions?

 

[dold]

When scheduling your design loads, do you provide a required moment, shear, axial, or whatever applicable scenario for each connection? Do you specify when loads occur in the same case or if you want the capacity to satisfy ASD or LRFD?

After answering 5000 RFIs asking essentially this question, I've started using a table with a whole bunch of notes. Table includes strictly nominal connection loads (load case basis) (DL, LL, LLr, SL, WL, EQ). I am very explicit about which code I have used to arrive at these loads, and I am very explicit about which code is to be used by the connection designer to combine the loads to arrive at the design load. This has reduced RFIs from connection designers and OWSJ designers, so that's good. Think of it as "if I were designing this, what information would I need?" As to which methodology they choose...it's kindof a grey area to be honest. A lot of folks I've worked with are adamant about using EITHER ASD or LRFD throughout the entire design, bar none. However, in my opinion, it seems that the connection designer should be able to choose which methodology will provide the most economical design. After all, that's kindof the whole point of moving to LRFD, no? But that's east coast market.

West coast / Rockies, you're most likely in SDC D+ (or D-? however you want to think about it...), so most of your SFRS connections are "special". I.e., capacity based design in a sense. So (moment frame example), you've got a beam size, beam size says connection needs to be *this* strong. And that's pretty much that - delegated or in-house. It's pretty cut and dry once you get over the fact that a 2" thick Gr. 50 end plate isnt the end of the world.

I might as well size the bolts/welds and put it in a schedule with a typical detail.

That's how we'd handle most brace and moment connections in earthquake town. Kinda like I said above - beam size >> spreadsheet >> end plate thickness, doublers, continuity plates, bolts >> schedule with Y-axis = beam size, X-axis = column size >> schematic detail + schedule >> connections fabb'd. Sometimes it works and the detailer knows the program. Sometimes you get some dude who does 'metalwork' in his shop and has no idea what the hell any of that means so you've got to essentially just do it yourself anyway.

So. I dunno where I'm going with any of that. If I had it my way I'd design my connections and delegate stairs and metal studs and all that overly-involved and uninteresting stuff to someone else who'd also rather not be doing that.