• State Suggested Answer
  • Replies 4 replies
  • Answers 2 answers
  • Subscribers 111 subscribers
  • Views 3723 views
  • Users 0 members are here

OMF Connection to HSS Column

Nulukkizdin
Offline

The project I'm working on is in SDC D (so no R = 3), but we fall into the exception of 12.2.5.6.1 which allows us to use OMF for our situation. Due to architectural constraints, we're looking at using HSS columns with wide flange beams to make up the OMF. Per the FR moment connection requirements of AISC 341, we are basically limited to designing our connection to 1.1RyMp. Presumably, as long as our connection can withstand this amplified force, it doesn't really matter what type of connection we choose (unlike IMF or SMF where we would need a prequalified connection).

My question is this. I have my frame system modeled up in RAM Elements with HSS columns and wide flange beams. The HSS to wide flange moment connections available in RAM Connection do not support seismic provisions. Is there a good way to design the connection in Elements using RAM Connection that will increase the demand moment to be 1.1RyMp based on the beam framing into it? My only other option is to manually calculate 1.1RyMp and apply that as a hard-coded moment at the end of my beam so that the connection sees it. If this is the case, I'll need two models - one to design connections and one to design members/drift. Also, each time my beam size changes, I will need to manually update the expected demand moment.

Parents
  • 0
    You will probably just have to enter the capacity based load manually. If you try to use a special moment frame connection it will reject the HSS shape.
    Also, you might want to check into the availability of A1085 hollow sections where you are. These have a more restricted range of Fy values and as a result a smaller Ry.



  • 0 in reply to Seth Guthrie
    How exactly is the A1085 supposed to be better? That's what I'd been hearing, but when we looked at specifying it on a recent SDC D job, it seemed to come out with higher loads on SCBF's. For instance, for SCBF connection tensile strength per 341-10, F2.6c, with an HSS4x4x1/2 brace, I would use Ry*Fy*Ag. Assuming A500 Grade B, I would get 1.4*46ksi*6.02in^2 = 387.68k. Assuming Grade C with the 341-16 Ry values, I would get 1.3*50ksi*6.02in^2 = 391.3k. But using A1085, with the 7% bigger Ag value, I would get 1.25*50ksi*6.36in^2 = 397.5k. Not only does the A1085 not come out any lower, it appears to be coming out about 2.5% higher on a capacity-based design. The 11% drop in Ry seems to be more than offset by the 8% increase in yield strength and removal of the old 0.93 thickness factor that was incorporated into the area. Am I missing something here?
Reply
  • 0 in reply to Seth Guthrie
    How exactly is the A1085 supposed to be better? That's what I'd been hearing, but when we looked at specifying it on a recent SDC D job, it seemed to come out with higher loads on SCBF's. For instance, for SCBF connection tensile strength per 341-10, F2.6c, with an HSS4x4x1/2 brace, I would use Ry*Fy*Ag. Assuming A500 Grade B, I would get 1.4*46ksi*6.02in^2 = 387.68k. Assuming Grade C with the 341-16 Ry values, I would get 1.3*50ksi*6.02in^2 = 391.3k. But using A1085, with the 7% bigger Ag value, I would get 1.25*50ksi*6.36in^2 = 397.5k. Not only does the A1085 not come out any lower, it appears to be coming out about 2.5% higher on a capacity-based design. The 11% drop in Ry seems to be more than offset by the 8% increase in yield strength and removal of the old 0.93 thickness factor that was incorporated into the area. Am I missing something here?
Children
  • 0 in reply to Jason McCool
    I was recalling a presentation I saw from Brad Fletcher, a representative of Atlas tube. When he mentioned the reduced Ry value I took note as potential for savings. In your example, I guess the thinking would be that the original size 4x4x1/2 might work as something thinner, e.g. A1085 HSS4x4x3/8, for strength and thus there would be that extra material savings in the member as well as the connection. If you are looking at the same nominal size then you are right, there is little to gain in connection design. He also noted that the yield stress on HSS tubes is usually about 55 ksi in reality, even if you specific A500 Grade B material. You might look him up or check with anyone at the Steel Tube Institute.



  • 0 in reply to Seth Guthrie
    Yeah, I ran into him and Kim Olsen from STI at the NCSEA Summit a couple of months ago. I'll check into that. She'd done an article a while back about how all the domestic HSS was produced as Grade C (50ksi) whether you asked for Grade B or not, so yeah, 55 is probably about right for actual MTR tensile coupon results. I picked that size as an example because it was one that works for seismic compactness requirements. We've run into problems before where we're in a vicious cycle of competing parameters between brace slenderness vs. width to thickness ratio. So a lot of times we can't go down on our thicknesses because of the dimensional restraints with global slenderness vs local wall buckling. What's interesting is that a size like 5x5x3/8 is no longer eligible as a SCBF brace with A1085. Table 1-5b in the 2010 SDM shows it as valid with Fy=46ksi. Bump it up to 50ksi, and it no longer meets the "highly ductile" criteria (barely). I just wonder if what appeared to be a big improvement initially (A1085) ends up getting lost in the details of our currently convoluted design requirements...
Communities
Support and Services
Training and Learning
Social Media