Tie beam moment with wall above and below

[Marc Rogue]

Hello guys,
I have pondered on this question before but never really tried to give it any answer. I typically work on multi story buildings no more that 5 stories, I have some design experience and just about to take the PE exam.

My questions is more of a theoretical one. in my experience I’ve mostly design moment frames with beams in all directions of the slab panels. My questions is if moment is directly proportional to curvature (I understand these curvatures are very small) how can we say that floor beam seating or sandwiched between walls experience moment and if they do so how can we justify the values obtained from the frame analysis.

I mean, some of the beams I design are mostly limited to a deflection of L/240 or smaller deflections. On a 15’ spam that’s a deflection of 3/4” and there isn’t a chance in hell any blockwall is allowing the beams to deflect that much.

I can see how edge beam would work as torsional members but I can only see interior beams as bearing members at best.

I’m of course not debating a long history of structural engineering here, I’m just wanting to get an understanding.

 

[HTURKAK]

Short answer ;If you have mostly designed moment frames and if the frame is constructed first ,and then the non load bearing walls constructed, the beams ( also columns ) will experience moment ..
If the walls are load bearing walls and the beams together with slab is casted on the load bearing walls your argument would be OK .
Another case , if the walls are continuous to the foundation, and if the wall restrained ( wedges placed at the gap btw top of wall and bottom of beam and the gap grouted ) the beam will experience less moment .


Deflection calculation is time consuming IMO , we should say deflection prediction rather than calculation. It should be limited to avoid cracks at walls and tiling etc..

Not to know is bad;
not to wish to know is worse.

NIGERIAN PROVERB

 

[Greenalleycat]

Assuming I'm understanding correctly your question is, what are the true demands on the beams in a moment frame that is infilled above and below? In which case I agree with you that there are basically none as the masonry will take most lateral and gravity load due to stiffness both in plane and axially.

There is some merit in saying construction sequencing will matter for the gravity loads as the masonry wall will be built after some of that has already been in place so you may need to consider that in allocating your gravity demands. Also you'd want to be very sure that there is goos bearing at the top - if the masonry wall is built first then the beam is poured on top then there will be solid bearing, but if the wall is built later, how will they reliably grout the top course to the underside of the beam?

There's whole area of investigation into infilled moment framed as the infilling massively changes their lateral performance so if you're in a seismic area you must look into that

 

[Marc Rogue]

Short answer ;If you have mostly designed moment frames and if the frame is constructed first ,and then the non load bearing walls constructed, the beams ( also columns ) will experience moment ..
If the walls are load bearing walls and the beams together with slab is casted on the load bearing walls your argument would be OK .
Another case , if the walls are continuous to the foundation, and if the wall restrained ( wedges placed at the gap btw top of wall and bottom of beam and the gap grouted ) the beam will experience less moment .

You guys both make good point, construction loads and dead loads would be a factor depending on the construction method. From my field experience at least in my area contractors would build footing/column, lay the blocks unto the the bot of beam elev. and then pour the beam and left over column. These procedure was what led me to believe these tie beams literally serve very little purpose other than as torsional members.

 

[KootK]

My questions is if moment is directly proportional to curvature (I understand these curvatures are very small) how can we say that floor beam seating or sandwiched between walls experience moment and if they do so how can we justify the values obtained from the frame analysis.

To my ears, it sounds as though you understand this issue perfectly. It is a fairly common occurrence that real world construction practices often result in ostensibly non-structural stuff getting packed in amongst the structural stuff such that:

1) The non-structural stuff winds up providing structural support and;

2) The structural stuff winds up not being taxed as anticipated because it's prevented from developing the displacements that beget stress.

To some extent, this begs the question "why have the structural stuff if it's not getting activated?". To that, I would answer:

3) As with seismic, sometimes the non-structural stuff trying to be structural causes problems that need to be addressed via redundancy or detailing that isolates things appropriately.

4) Examined in detail, the load paths represented by the non-structural stuff often tend to have holes in them.

 

[BAretired]

And sometimes the non-structural stuff is removed by someone who thinks "who cares?" it's non-structural stuff.

 

[Marc Rogue]

To my ears, it sounds as though you understand this issue perfectly. It is a fairly common occurrence that real world construction practices often result in ostensibly non-structural stuff getting packed in amongst the structural stuff such that:

Your comments ties right into my most pressing concern. If the structures I design are not working as intended due to the method in which it’s being constructed then I worry it should also be up to me to tell the contractor how to construct the structure as well meaning I should create a schedule or something of the sort instructing the sequence of construction as I can not possibly expect the contractor to know the intent of the design.

I know it’s probably second nature to some of you but this stuff really concerns me as my jobs have started to get bigger and bigger and the load demands are more and more critical.

 

[KootK]

It is a structural engineers duty to suitably protect the non-structural parts of ther buildings from damage that would be caused by anticipated deformations of the primary structure. Some strategies that we use for that include:

1) Isolating the non structural elements. Slip tracks at the top of cold formed stud walls and stuff like that.

2) Designing the primary structure to be suitably stiff that it does not tax the non-structural elements harmfully. L/600 etc, absolute deflection limits etc.

3) Taking advantage of construction sequence where appropriate, acknowledging that non-structural elements only care about the deflection that occurs after their installation.

4) In rare situtaions, designing the non-structural elements to be what I'd call pseudo structural. In many parts of the developing world, I feel that CMU infilled masonry buildings create this situation. And that sometimes begets the seismic problems.

In fact, I consider protection of the non-structural building elements from damage induced by the primary structure to be one of the key aspects of structural design. Life safety trumps all else ethically but it's usually serviceability problems that land in hot water with your clients or -- worse -- they're lawyers.

I worry it should also be up to me to tell the contractor how to construct the structure as well meaning I should create a schedule or something of the sort instructing the sequence of construction as I can not possibly expect the contractor to know the intent of the design.

I agree that there are some situations where specifying the sequence of construction makes sense but, for common / conventional buildings, I would personally avoid that. For the most part, contractors use the sequence that makes sense to them and any interference with that can be expected to be either a) ignored or b) add cost. For common buildings, I only rely on sequence of construction where I'm confident that it will play out the way that I want it to without any active specification on my part.