Economy in Steel Framed Multistory Structures - Moment Connections or Pinned Connects

[human909]

Here is a question I thought I'd throw out to the masses. I'm after what people find as the most effective/economic solution to a structure of the form:
-Steel framed, open section beams and columns
-Braced frames for lateral restraint
-Multiple stories (let say 3-8 stories)
-Industrial setting envisaged so less architectural constraints.

Have you found benefits in using moment connections for the primary beams? Given there is already lateral restraint the benefits of the moment connection mostly lie in better economy of the relevant beams. A simple example is shown in the plan view below where all the beams running north-south are moment connections.

(Obviously most of use know the trade offs between moment connections and simply supported. Moment connections generally make more efficient use of your members but connection detailing can be more difficult both for engineering and cost of fabrication and erection. )

 

[r13]

I would provide moment connection for all beam-column joints, and simple shear connection for beam-beam joints, for ease of construction.

 

[jayrod12]

I'm in the all shear only connections if you've got a braced frame. This is assuming there's adequate bracing to have some redundancy in the lateral system. If losing a single brace would be detrimental, then I agree with retired13.

 

[SlideRuleEra]

Industrial setting envisaged so less architectural constraints.

Shear connections. When it's time to added unanticipated equipment, or modify/adapt the structure for new uses, best to have a straightforward design where there is a chance that existing members have at least some reserve capacity. Economical first cost is secondary.

http://www.SlideRuleEra.net

 

[MIStructE_IRE]

Pin - pin for me all the way.

 

[human909]

Looks likes a most strong responses in favor of keeping it simple.

I've mostly seen simple connections in such structures but I did see somebody else design in some moment connections for the primary beams in something I was involved in. Recently I've designed a few structures where significant moment connections were necessary. Now I'm back looking at a big braced greenfield plant and considering whether the deflection and strength advantages of a moment connections can bring in benefits.

I would provide moment connection for all beam-column joints, and simple shear connection for beam-beam joints, for ease of construction.

That is what I'm thinking, you make the most of the primary beams. For sure beam-beam-joist connections to be simple, I don't want to make my life or the erection crew hell! (Though multistory and multi bay buildings would required pretty tight tolerances with moment connections or your erection crew will hate you. Shims can work but who wants to deal with that!) The latter is a good reason to be wary.

 

[HTURKAK]

I will suggest the use of in-between case for beam - column joints ; semirigid connections. The semirigid conn. modelling creates lighter frames. The reason is that , the overall maximum moments in beams decrease by comparison with the moments of rigid frames for hogging moments and decrease the sagging moment which is max. for the simple connections , thus smaller sections can be used.

However the semi-rigid connections cause a large increase in the frame drift if the bld is not braced. I thing this thread discuseed in past several times , one is , thread507-461579.

I found the pictures from web search which is useful to see the concept.

 

[r13]

For industrial setting, heavy uneven loads are anticipated, rigidity of floor system is paramountly important.

 

[human909]

I will suggest the use of in-between case for beam - column joints ; semirigid connections. The semirigid conn. modelling creates lighter frames.

That looks great in theory but creating a semi rigid connection to perform in a reliable, quantifiable and code justifiable manner seems quite difficult. But that is something I'll look at for my own curiosity, but I don't see the benefits to complicate things too much for this type of application.

For industrial setting, heavy uneven loads are anticipated, rigidity of floor system is paramountly important.

But that doesn't require rigid connections as you surely must know. Though if your point was that a fully rigid moment connection gives you better a more rigid flooring with less deflection and less vibration for a given member size then I agree with you. I've seen other constructions of this type that have entire floors shaking to an intolerable degree due to flexible members supporting large vibratory sifters.

That said I wouldn't say that flooring systems of many industrial settings are need to be particularly more rigid than other settings. In fact in many cases floor wide vibrations are often more acceptable and tolerable than say a ballroom or a conference room. (Though "industrial" applications obvious vary immensely a drop forge has different needs to a silicon chip manufacturer)

 

[r13]

...less deflection...

That's pretty much my concern on the existing floor framing layout (30'x45' per bay?). I'll change my mind if you can add a few more columns.

 

[human909]

That's pretty much my concern on the existing floor framing layout (30'x45' per bay?). I'll change my mind if you can add a few more columns.

I think we are on the same page. I just initially misunderstood your comments.

Regarding that picture. That is just an example I had on hand. It is smaller 30'x21' building, nothing too complex.

The new building that will be designed from scratch will be 50'x85', 90' high. There will be 5or6 by 3 column grid most likely. But design hasn't started. It will be similar to a previous job we did but we exact requirements and design still to be determined.

 

[r13]

Understand, I wasn't very clear in my response.

BTW, although you didn't mention floor slab, but I guess you would utilize composite design to lighten the steel framing a little and keep deflection in check, in such a case, all connections are preferably be made of simple shear connection.

 

[human909]

Understand, I wasn't very clear in my response.

BTW, although you didn't mention floor slab, but I guess you would utilize composite design to lighten the steel framing a little and keep deflection in check, in such a case, all connections are preferably be made of simple shear connection.

Steel plate flooring. Penetrations everywhere as there the process flow is half gravity driven, hence the height.

Since you are curious, the example I gave earlier of the 50'x85', 90' high building that is a rough guide columns were mostly 310UC and beams 360UB or 410UB. Or in you language W12x79 columns and S15x42.9 beams (roughly).

All this is going to get redesigned from scratch. But I just thought I'd ask the question regarding the use of moment connections in braced structures. I think the previous building excessively conservative in many areas. But once I start plugging in loadings I'll get my answers.

 

[johnie134]

in most cases, braced frames are economically more suitable.

 

[human909]

To answer my own question.

Running a few designs myself and fleshing it out a bit more it would seem that the benefits from moment connections are unlikely to outweigh their increased costs for shorter spans that are more heavily loaded. Where deflection/vibration is heavily dominant such as longer spans which are lightly loaded then the benefits for moment connections present themselves more strongly. A semi-obvious conclusion when you think about it.

I'm sure there would be plenty of sources out there that repeat similar conclusions. But I thought I'd ask the community and there is always benefits to examining scenarios yourself.

 

[r13]

it would seem that the benefits from moment connections are unlikely to outweigh their increased costs for shorter spans that are more heavily loaded.

I think the conclusion is quite accurate. I personally does not have a "rule of thumb", but many have used one when doing preliminary layout. Span length is always one of the prominent factors in framing system selection.

 

[CANPRO]

Fabricator perspective here.

The cost of the connections can't be understated - you're only going to get a lighter beam if you draw significant bending moments to the end connections. When looking at the material savings in your beams, are you also considering the additional weight being added to your columns? Your columns are going to see additional bending moments especially with unbalanced loads and/or girder spans.

I'm of the opinion that using fixed end connections for girders is likely not an economical solution for reducing deflection/vibrations regardless of the span and load conditions. Just up-size the beam.

Moment connections are going to add cost and time to the construction schedule right down the line from connection design, detailing, fabrication, and erection. The erection time probably being the biggest cost factor. Not to mention that moment connections are more complex than simple shear and therefore more prone to design/drafting/fabrication errors - which all lead to additional site costs.

It would be very helpful if you knew a local fabricator or two you could reach out to and get some preliminary estimates on a couple of options.

 

[Aesur]

I agree with CANPRO, those connections can get complicated and add significant cost and time to a project as well as most likely require a field welder for a longer period of time. I am of the opinion that the moment connections would add significant costs compared to pinned connections.

 

[Tomfh]

Pinned and braced where possible.

 

[human909]

Fabricator perspective here.

The cost of the connections can't be understated......

I completely agree. And I've normally jumped straight to the same point, my first job was for a steel fabrication firm.

I only started to wonder because I was involved in a braced structure last year that used moment connections and that started me wondering whether it makes sense. I've also designed several 'drive through' moment structures recently supporting thousands of tonnes. But those structure had no room for bracing. Despite moment connections they were actually surprisingly cheap to build in a $/tonne perspective. Heavy beams and columns!

It would be very helpful if you knew a local fabricator or two you could reach out to and get some preliminary estimates on a couple of options.

Good advice to any engineer. Though in this case I've already dismissed the idea.