Moment Connection?

[SteelPE]

I have a question that is probably very simple for some.

I have been asked to design some moment connections for a fabricator. I pretty much know how to do it, however, applying the loads to the column for some reason seems like voodoo to me.

The EOR was kind enough to give use the beam end moments to apply to the columns (see attached). However, he was not nice enough to say whether or not the moments are based upon LRDF or ASD load combinations (something we have inquired about). I suspect they used ASD combinations. The columns shown in this sketch are single story columns therefore the columns do not continue to another floor.

Applying the moment to the end columns seems simple enough. However, applying the moment to the middle columns for some reason I just don’t get. Now given the sample arrangement, I would apply 250ft-kips to each side of the column (similar to adding 500 ft-kips to one side of the column). However, there is no way a W12x79 is going to be able to take this 500 ft-kip loading. This is what I don’t get. Why design the connection for this load if there is no way the members can take these loads.

Is adding the reactions together appropriate for the center column?

 

[ztengguy]

wouldnt the 250K be in opposite directions, thus 0 local moment?

I would suspect the EOR should have checked the moment in the column, and run a live load one side case or something like that.

 

[Lion06]

My first question is, is this a moment frame and these lateral moments or is this just a gravity frame and the moments are gravity?

It sounds like it's a lateral frame since you're adding the moments at the interior columns.

Here are just a couple thoughts. The 250 k-ft moment is probably a combination of gravity and lateral forces. As such, you'll never have a full 500 k-ft cranking into the top of the column, because the gravity portion of the moments are counteracting each other. I was going to suggest requesting the moment for each load case, but you don't really care about that (other than the question posed here) for your design, because the connection doesn't care if the moments are additive or counteracting each other.

 

[ishvaaag]

Well, kind or not, the issue needs be cleared. First, if moments are service level combinations, second, sign of the applied moments to the sides. If you (the firm) want not to be in business disruption by making such kind of statement, you need to access to the project documentation and from it and use rebuild the analysis to ascertain what you have in hands. But normally your boss in the alternative will prefer asking loud.

 

[SteelPE]

ztnegguy

My thought process is, who is to say the moments are not from a lateral load case. Also, I'm not sure if the column/panel zone really cares.

Lion,

I imagine these are lateral frames that take gravity forces. I don't really see any other LFRS in the structure. There are also a few other peculiar notes on the drawing that we have asked clarification on.

I don't really want to deal 900 load combinations. I'm happy just dealing with the information they provided to us.

 

[nutte]

You need to know if the moments are ASD or LRFD. Nobody can tell you this other than the EOR.

The EOR also needs to tell you if the moments are additive (making a column moment of 500 k-ft) or not (resulting in no moment being transferred to column). It does matter for design of the column panel shear zone.

If the moments are additive, and the column can't take 500 k-ft, you could design the column panel zone for a moment equal to the column's capacity, assuming the EOR won't give you the actual moment being transferred into the column.

 

[SteelPE]

nutte,

If the moments are additive, then wouldn't this have an effect on the panel zone? I would agree if they are strictly due to gravity loads and the spans are equal then the column would see no net load. For this case (equal spans) I would only need to be concerned about the continuity plates.

However, if you look at the sketch I provided, the center beam span is smaller than the outside beam spans. Therefore, I believe we would still end up with some amount of gravity moment being cranked into the column. Panel zone and web stiffeners would need to be designed accordingly.

I also know that sometimes the Seismic Provisions will govern the design making for weird connection designs. Working with the building code, these are usually provisions that we avoid knowing the problems they can cause with the connection design. This is another bit of information that was missing from the drawings but was asked to be supplied by the EOR.

 

[Lion06]

steelPE-

The column panel zone shear does care. If the moments completely counteract each other then there is no column panel zone shear.

Additionally, even if this is a lateral frame, not ALL of the moment is coming from lateral forces. The gravity portions of the moments WILL counteract each other and NOT get cranked into the top of the column.

I don't think that you need any additional information to design the actual connections, but if you're concerned about the column, either ask the EOR the question or ask for a breakdown of moments by load case, not load combination. The breakdown by case will be far more helpful.

I understand your concern and comment about who is to say the moments aren't lateral moments. But the truth is, you don't know.

 

[SteelPE]

Lion

Would the gravity load cases be equal and opposite if the span was the same on each side of the column? I can see some moment being "cranked" into the column for unequal spans. I must be missing something.

I will agree that getting load case forces would be a little easier but this is dependent on there being just the basic cases (Dead, Live, Snow, Wind and Seismic). If they break the lateral force cases into 30 cases then it would be a slightly bigger pain.

 

[Lion06]

No, you're correct, unequal spans will likely result in different gravity moments on each side of the column, so some will get cranked in, but not all of it. The important thing is that what doesn't get cranked in is doubled. What I mean by that is if the lowest gravity moment on either side of the beam is 50k-ft, that's 100k-ft of moment that isn't getting cranked into the column because it's counteracting 50k-ft on the other side.

When I've been asked to provide connection force information for a condition like this, I've given DL, LL, WL (worst case for connection, not 30 cases for each), and E (same as WL).

 

[spats]

Simply put, the unbalanced moment in the beams gets tranfered into the column.

 

[connectegr]

The moments given are probably envelop forces. The lateral moments and gravity moment either add or subtract to determine the column web shear and web doubler requirements. The flange forces and stiffener requirements are determined by a combination of the gravity and lateral loads. If lateral moments are given then this is clearly a moment frame and the forces are not equal and opposite, as in a cantilever. Since the forces are envelop, their combination probably does not represent the forces used to size the column. I deal with this everyday. Good luck getting the most practical response from the EOR.

http://www.FerrellEngineering.com

 

[SteelPE]

connectegr,

I was hoping you would have some input. I realize this is something you probably deal with on a regular basis. In your experience, how helpful is the EOR with providing you the information you need? We have had an RFI out there for a few days asking about code errors, seismic criteria, and method of moment development but have yet to hear a response.

I can design the connection for the forces shown, however, there is no way it is going to be economical, and there is no way my client is going to like it.

We may just end up passing on the job, but I would still like to learn what should be done for future reference.

 

[nutte]

The EORs I've dealt with range from being very helpful to downright antagonistic. Sometimes I come across one who knows exactly what I'm looking for and is willing to provide that information. Usually, they take the easy way out and tell us to use 500 k-ft for the column. If you point out that the column can't take it, they may say "OK, use the column capacity," and they may say "sorry, use 500 k-ft."

I can design the connection for the forces shown, however, there is no way it is going to be economical, and there is no way my client is going to like it.

True, and the EOR may not care one bit about this. Having a cooperative EOR makes a huge difference.

 

[SteelPE]

Well, at least I know I was on the right track. Whether or not it made any sense was not a decision I could make.

The question I had also involves AISC 341. However, I believe that even the connections I have seen done under AISC 341 also limit the capacity of the connection to the column strength.

 

[connectegr]

Some EOR's are very helpful, but most cases provide less information than you have already received. My typical RFI for your case... "Please verify that no web doublers are required, or provide the lateral and gravity moments to determine the required reinforcement" I also provide the attached detail for determining the column shear. Some structural analysis programs will check the local column shears and verify the need for web doublers. If the EOR states that no doublers are required, the direction of the moments does not matter. It would be nice to have the actual moment forces, but that is rarely given.

Ideally the EOR works with you. Bill Thornton has said that reinforcement of columns, material and labor, can equate to 79 lbs/ft in column weight. In my experience, and with fabricator input, that difference is comfortably 50 lbs/ft. That leaves big room for adjusting the column size, to provide a "clean column" and big project savings. In most cases only a little additional weight can do the trick.

In the EOR's defense, providing adequate information for a safe connection is his/her only responsibility. Conservative is certainly safe. There is rarely any additional fee for working well with the connection designer or fabricator. I have rarely convinced a fabricator to offer a credit or reimbursement to offset the EOR's time. Actually I have only seen the fabricator pay a fee to the EOR when the alternatives were massive field repairs.

Additionally, these conversations generally work best engineer to engineer. No one likes mountains of RFI's. The fabricator and contractor's translation of your questions or comments can do more harm than good. I always assure all parties that a tel-con will be provided for all solutions reached in our discussions.

http://www.FerrellEngineering.com

 

[SteelPE]

I realize the extremely large cost when deal with doubler plates. This is why I always try to avoid them when I design a structure. However, in this instance help from the EOR is going to be required.

This particular fabricator has specifically excluded web doublers in their quote because of the added cost. I don't know how that is going to work out for them going forward. However, it's not a fight I have to deal with.

So you have actually had success in asking the EOR if doubler plates are required? I would think they wouldn't even answer the question.

I do remember a job from long ago where the EOR asked for the connection to develop the full capacity of the beam. The connection was two 24x84's framing into a W12x53. The W12x53 wouldn't even come close to developing the required moment. When pressed the EOR said that we only needed to develop 2x the capacity of the column (because the column continued on up through the connection).

 

[sandman21]

The same exact moment for each column, I wonder how they got that. lol

OMF AISC 341 section 11.2A requires 1.1RyMp from the column or beam for connection designs