Choosing between braced frames and moment frames 2

[abusementpark]

When you have a typical, multi-story (say 3-5 stories), steel-framed, commercial office building, what do you generally find to be the most efficient/least expensive lateral force resisting system?

Do you always try to make braced frames work if the architecture allows for it? Or are there some scenarios where moment frames might be a better option, even if braced frames are possible? I know that each job is unique, but what is your general process for making this decision?

 

[steellion]

The conventional wisdom is that braced frames are always more economical than moment frames with everything else being equal (by a factor of 2 or 3). We always use braced frames wherever possible, even the result is a non-typical brace configuration to work around the architecture. I've never actually run the numbers, but you need to seriously upsize the members in a moment frame to get the same drift performance that you get from a braced frame.

 

[hawkaz]

Shear walls- if you can get enough wall that stacks

If you are limited to moment frames vs braced framed, I agree with Steellion
One of the main disadvantages with moment frames is deflection- you will most likely need to fix the based to keep this in check

 

[DST148]

Ref ASCE 7-05, Chapter 12: R for OSCB is 3.25, R for OSMF is 3.5, R for steel systems not specifically detailed for seismic resistance is 3.
We do mid-rise buildings, 10 to 15 story high, and typically fall in SDC B or C. Since R for the above mentioned lateral force resisting systems is comparable, we have used concentrically braced steel frames not detailed for seismic resistance. For OSCB and OSMF, seismic detailing per AISC 341 is required.

 

[abusementpark]

We generally use braced frames. But the thing I hate about braced frames is how all the shear load gets concentrated at discreet locations in the building structure. Also, you end up with a lot of high uplift reactions since you don't engage very much of the building's weight in resisting the overturning. There always seems to be a large disparity between what we need for foundation at the lateral columns as compared to the gravity columns.

I've always envisioned that the foundation system would be a lot more efficient if you used moment frames. But, I guess that doesn't come close to outweighing the costs associated with upping all the member sizes to keep the drift in check.

 

[SteelPE]

abusementpark,

I just ran through this exercise on a building. It was a small building (3 stories) but on a site class E. The preliminary foundations are ridiculously large for such a small building. The architect barely allowed any place for bracing. I quickly ran through the numbers for the system using moment frames and found that the steel sizes went through the roof. In the end I decided to keep the brace frames with the ridiculous foundations.

If the owner or GC have a problem with what I came up with well I have plenty of suggestions to make things more reasonable..... they just need to live with some of my changes.

 

[WillisV]

See attached for a comparison/cost study. Braced frames are almost always more efficient if possible. Moment frames 2 to 4 times as much.

 

[SteelPE]

Willis

They suggest fixing the base of the columns to control drift in the first story.... everyone on this forum always says to never "fix" the bases because you never get a truly fixed base at that increases the cost of the system to much.

 

[WillisV]

@SteelPE - I generally do fix the base of moment frames so I guess I am the exception to everyone. You certainly pay a penalty in larger foundations, baseplates, and anchor rods but in my experience the reduction in column and beam sizes required to meet drift limits is worth this penalty.

Another item to note is that you can gain a good bit of siffness in the frame by taking into account composite action at the beams if you have it. Most people don't because it is easier to ignore it and most commercial programs (Ram for instance) ignores it, but it is there if you want to put forth the effort and properly account for negative moment regions etc.

 

[abusementpark]

I just ran through this exercise on a building. It was a small building (3 stories) but on a site class E. The preliminary foundations are ridiculously large for such a small building. The architect barely allowed any place for bracing. I quickly ran through the numbers for the system using moment frames and found that the steel sizes went through the roof. In the end I decided to keep the brace frames with the ridiculous foundations.

Did you do a one-for-one substitution of the moment frames for the braced frames? Or did you make all the connections moment-resisting?

 

[SteelPE]

No, I re framed the building using what I thought was the best locations for moment frames. I used partially fixed bases and had ended up with a huge increase in the column size and deflections still wasn't even close. I did a bunch of work for a steel fabricator at my first job and I knew the moment frames were not going to be cost effective. So, rather than spend a few more hours refining the frames I ditched the idea and kept going.

Part of what was causing problems was frame drift. This particular structure was clad in brick so that didn't help at all. I stayed away from fixing the bases because of what others have warned me about and because the building slab was 4' above grade (to pull it out of a flood plain) making the footings very deep and pilasters very tall.

 

[abusementpark]

I have never gone through this exercise, but I would have thought moment frames would be most viable if you made every frame line moment-resisting. That way, the way you would limit the work each frame would have to do, and maybe the sizes wouldn't be so significant. And you would be using the full depth and weight of the building to resist overturning. But again, I have never run the numbers to see.

 

[WillisV]

@abusement park. The costs of creating full moment connections at every frame line would be astronomical. This line of thinking (well distributed moment resisting system at most connections) though is the primary advantage of a partially restrained (PR) building, where connections with some moment resistance, typically top and bottom angles, are used throughout the building. This reduces the costs drastically from a true full moment connection and achieves the same goal. The main problem is they are a pain to design, and we generally do not get enough fee to deal with it.

 

[abusementpark]

The costs of creating full moment connections at every frame line would be astronomical. This line of thinking (well distributed moment resisting system at most connections) though is the primary advantage of a partially restrained (PR) building, where connections with some moment resistance, typically top and bottom angles, are used throughout the building. This reduces the costs drastically from a true full moment connection and achieves the same goal. The main problem is they are a pain to design, and we generally do not get enough fee to deal with it.

What about the costs of the extended end plate bolted moment connection, instead of the traditional connection with field welding of the flanges? I've wondered how much more expensive that connection is compared to a typical shear connection, since you eliminate the field welding.

 

[slickdeals]

Or side plate moment connections.

It’s no trick to get the answers when you have all the data. The trick is to get the answers when you only have half the data and half that is wrong and you don’t know which half - LORD KELVIN

 

[steellion]

@WillisV, I consider moment frames fixed at the base and design the foundation and connection at the base accordingly for the same reasons that you mention. I'm confident you are not alone. I do not consider any additional capacity from the concrete in a composite system. I'm not sure how you can consider the concrete when a portal frame analysis indicates +M on one side of the beam and -M on the other side. Do you add bars in the concrete over the moment frame and consider the rebar in tension?

 

[WillisV]

@steellion. You certainly can do that if you want additional stiffness in the negative moment region, but in general its easier is just take into account the increased stiffness due to composite action within only the positive moment area. I'm not talking about additional capacity here - just additional stiffness to help with drift at moment frames. See Commentary I3, equation C-I3-2 (13th Ed. p16.1-309) for an extremely simplified way of doing that. For lateral loading as stated in the Commentary you end up with an average Iavg of 0.5*I of the steel alone + 0.5*I of the composite section moment of inertia to use for lateral design. For programs that allow it this is fairly easy to implement as you just use a stiffness modifier on the bare steel equal to Iavg / Isteel for your drift calcs.

 

[steellion]

@WillisV, interesting stuff, thanks for the tip!