Fire resistance of 8" thick concrete column 1

[milkshakelake]

Assuming slenderness and strength is okay, is there a minimum width of an 8" thick concrete column for fire resistance in NYC? I looked up and down through codes and couldn't find anything.

A past coworker years ago told me that an 8" column has to be either 24" or 36" long minimum due to fire codes; I can't remember the exact number. I dismissed it at the time because I never thought about using an 8" column. Now that I have my own practice and I'm working with owners that are pushing for this, I wonder what the requirements are. I'm also afraid that once the aspect ratio is more than 2:1, it becomes a shear wall and I can't ignore it in lateral design.

 

[slickdeals]

I am not familiar with the NY Building Code, but for a column with 8" thickness you can only get 1 hour fire rating. See Table 722.2.4 (IBC 2015)
However, footnote (a) suggests that a column dimension 8" x 36" can achieve up to a 3 hour fire rating, which is typically what's required for columns in a Type 1 construction.

 

[milkshakelake]

slickdeals - Thanks for the reply. That's exactly what I was looking for, and I'm sure it also applies to NYC. Thanks! As a further question, would you consider such a column a shear wall?

 

[slickdeals]

I didn't know if there was a definitive answer on this, but ACI 318-14 now clarifies this.
Refer to definition of walls in Section 2.3 of ACI 318-14.

Wall – a vertical element designed to resist axial load, lateral load, or both, with a horizontal length-to-thickness ratio greater than 3, used to enclose or separate spaces.

According to ACI 318-14, this 8" x 36" element would be considered a "wall".

However,personally, I would design it as a column and adhere to the detailing requirements of a column and not a wall.

I am sure others here will chime in.

 

[milkshakelake]

I figured that lateral forces go to the stiffest members (i.e. shear walls), so these columns are okay. In the event of an earthquake, a shear wall would carry most of the forces. Even if the columns carried some of the lateral forces, if they failed, the shear walls would take up the slack.

I saw a post about crazy amounts of walking columns, and how they would unintentionally attract seismic forces. That's true, but for "rectangular" small buildings like what I design (up to 10 stories without crazy walks, no torsional or stiffness irregularities), it wouldn't apply.

 

[KootK]

That's true, but for "rectangular" small buildings like what I design (up to 10 stories without crazy walks, no torsional or stiffness irregularities), it wouldn't apply.

I get where you're coming from but do not share your confidence. Even without the walks, your lateral loads will distribute based on stiffness. Wind included. Back when I got started, there seemed to be a rule of thumb that you could ignore the lateral contribution of the columns if their combined stiffness added up to less than 1/6th of the shear wall system stiffness. Where that came from and whether or not it's got any real merit, I couldn't say.

Even if the columns carried some of the lateral forces, if they failed, the shear walls would take up the slack.

The question is this: if the columns "fail" for resisting shear loads, can they still be relied upon to resist gravity loads? Obviously, resisting gravity load is a rather important function of multistory building columns. One common issue comes about when wall-umns exposed to lateral drift develop high punching shear stresses at the wall ends and, potentially, generate punching shear failures that compromise gravity load carrying capacity.

I'm an entrepreneur, like you. While I'm not saying that you need to make a PhD thesis out of this, do take care not to be excessively cavalier in your decision making en route preserving fee and keeping your clients happy. If you've got ETABS etc, you should be able to make pretty quick work out of assessing the potential for trouble here.

 

[slickdeals]

Back when I got started, there seemed to be a rule of thumb that you could ignore the lateral contribution of the columns if their combined stiffness added up to less than 1/6th of the shear wall system stiffness

There is a requirement of what qualifies as a non-sway (braced) system in ACI 318. Refer to section 6.2.5 (ACI 318-14). The requirement is shear walls having 12 times the stiffness of that of columns.

 

[KootK]

Thanks Slick. I'd been curious about the fire ratings on thin wallumns too so you've educated me a fair bit here. I'm grateful.

 

[milkshakelake]

slickdeals - Thank you for that code. My shear walls clearly have more than 12 times the stiffness of the long columns.

KootK - I've read articles about punching shear failures, and it's scary. Usually flat slabs are okay in bending but punching shear always controls. I haven't read anything about lateral punching shear failures though. I'm not disagreeing with you. It's just not something I'm familiar with; if you have a case study, I'd love to read it.
I totally get what you're saying though. I will err on the side of caution and make sure that these long columns are okay in punching shear.

 

[KootK]

if you have a case study, I'd love to read it.

I don't have anything handy. However, it is this phenomenon that largely underlies the ACI 318 provisions that:

1) Prohibit the use of flat plate & column frames for use as the lateral system in seismic areas and;

2) Require the slab to column joints to be analyzed for drift compatibility in seismic areas.

I'm sure some thoughtful Googling would turn up the case studies that you seek.

My shear walls clearly have more than 12 times the stiffness of the long columns.

Do they have 12X the stiffness of all the long columns added together? Additionally, the stiffness of the shear walls needs to consider the height of the walls up to the level under consideration for flexibility whereas the columns should be evaluated at the story height. That will make a pretty big difference near the top.

Frankly, I'm not entirely sure that Slick has the spirit of this provision correct anyhow. It reads as though it is about classifying columns as non-sway which, although similar to what we're talking about here, is not quite the same in my opinion. I'll noodle on that.

I will err on the side of caution and make sure that these long columns are okay in punching shear.

A feel good trick that I've been using is to run ample, integrity, bottom steel across the short direction of the columns. That way, the slabs aren't going any place even if you do get a punching shear failure. You might have some unit owners that are unhappy about the appearance of their slabs but at least they'll be alive to complain about it.

 

[milkshakelake]

KootK - Thanks for the reply. I'm not in a seismic zone (NYC) so I never thought about that. In NYC, people use slab-column interaction all the time. Though they probably do it the wrong way; I found that typical column sizes (like 10"x22" or 12"x24") and typical slab sizes (like 8") will almost always fail the slab in punching shear if they are taking lateral forces.

Frankly, I'm not entirely sure that Slick has the spirit of this provision correct anyhow. It reads as though it is about classifying columns as non-sway which, although similar to what we're talking about here, is not quite the same in my opinion.

I interpreted slickdeal's comment in the same way. Now that I think about it, if you have a non-sway column, it won't experience too many P-delta effects but it can still be stiff enough to take lateral forces. This line of thinking invalidates hundreds of columns I've designed, so I hope there's something more to it that I haven't thought of yet.

Additionally, the stiffness of the shear walls needs to consider the height of the walls up to the level under consideration for flexibility whereas the columns should be evaluated at the story height.

Can you explain this further? So let's say I have a 2 story shear wall. The stiffness of the shear wall at roof will be cantilevered from the ground, whereas the stiffness of the columns are one story?

 

[KootK]

Can you explain this further?

Sure, although it sounds as though you've pretty much gotten the gist of it. I'll start by saying that it's something I've not entirely sorted out myself however.

1) The way that I've seen most engineers do this check is to look at the plan configuration of the lateral elements and columns, calculate the Ix values for each, and see if Ix_lateral_systems >> Ix_sum_of_columns. I feel that this is ridiculous for a building of any appreciable height because the columns and shear walls will behave fundamentally differently. Your shear walls will act like vertical cantilevers and your columns will act like shoddy moment frames with repeating inflection points in the columns (as opposed to cantilever)?. So it's an apples and oranges comparison in my estimation. It's quite analogous to the whole shear wall / frame interaction business shown below.

2) In my opinion, the right way to do this would be to truly compare lateral stiffnesses (load vs displacement) at any two adjacent floors. Something like this:

2a) Hold the 30th floor still and push the 31st floor past it 1" laterally.

2b) See how much load goes to each system based on the model shown below or something akin to it.

2c) Make your determination about whether or not the columns participate meaningfully in lateral load resistance on this basis.

Obviously, this would be a ton of work. Particularly so given that you'd need to look at each floor individually or groups of floors individually. Much easier to just fire up ETABS.

While I get the spirit of your two story example, it's uniquely inappropriate for this phenomenon. With walls that short, you're likely to have both your columns and your walls dominated by shear deformation.

 

[milkshakelake]

KootK - Thanks for the detailed response! I remember learning that in school, now I just need to brush up on it.