ACI 318- 19: Does ACI specify a minimum edge distance for a punching shear perimeter to be effective? 1

[jochav52802]

Good Day!

Does ACI specify the minimum edge distance required beyond the critical perimeter for that perimeter to be effective in two-way shear?

All I can find is Section-22.6.4.3, where they say that openings located less than 4h from the column face eliminate the effectiveness of the associated critical perimeter, but am unsure as to whether this is to be used for identifying general free edges as well.

Wondering if this 4h also applies to the outermost critical section when punching shear reinforcement is required. I don't see the CRSI Design Guide for Pile Caps performing any edge distance check on the critical perimeter, so wondering what is best practice here. Intuitively and practically speaking, at some point, when there's very little edge distance, I would think the adjacent critical perimeter would not be effective.

Many thanks in advance for your wisdom here!

 

[KootK]

Does ACI specify the minimum edge distance required beyond the critical perimeter for that perimeter to be effective in two-way shear?

They certainly have in the past and I'd be astonished if they don't still. My recollection is 5h.

 

[jochav52802]

They certainly have in the past and I'd be astonished if they don't still. My recollection is 5h.

Thank you, KootK,

Any chance you could take a look and tell me where to find that? All I'm seeing is the 4h in Section 22.6.4.3, but that's specifically for openings.

 

[KootK]

I took a quick look and couldn't find anything definitive fast enough. The language applied to openings kind of suggests that openings and free edges are treated the same. So 4d perhaps.

 

[jochav52802]

I took a quick look and couldn't find anything definitive fast enough. The language applied to openings kind of suggests that openings and free edges are treated the same. So 4d perhaps.

Thank you, KootK.

Kind of a bummer ACI isn't more explicit here; I'd think edge distance would be pretty important. Fig. R22.6.4.3 references free edges, but only references openings when it states the 4h requirement, so nothing definitive on the free edge front.

 

[KootK]

Agreed. Unless you can scare up something that specifically says otherwise, I'd go with 4d.

It might also be worth checking out how software is treating this. I don't love using that as "definitive" but, at the same time, those folks have a significant interest in getting it right. Just gotta keep an eye on them being to conservative with CYA.

It might take a day or two but I'd be surprised if someone here can't add more clarity to this. @Celt83.

 

[jochav52802]

Thank you, KootK.

I appreciate that; I'll see if there's any wisdom from the software people.

Just for your reference, in regard to the smaller sized standard pile caps shown within the CRSI Design Guide for Pile Caps, many of them would have to get significantly larger to meet the 4h requirement, so it seems they're not following it, (e.g. the 4 pile cap is called out as 5'-6" square, but with it's thinnest pile cap of 31", 4h would require a 21'-8" square pile cap, which is obviously unreasonable).

 

[Celt83]

My understanding is there is no limit in ACI for which a free edge does not need to be considered.

In my experience for ACI software tends to follow ACI 318-19 figure R22.6.4.3 (bottom left) and extend the side perimeters in a straight line to the free edge. You usually give the software a radius to search withing for critical sections and the software designs for each complete shear perimeter it can form.

I haven't tested this myself but imagine something like 10*h probably gets you in the ballpark where the complete interior column perimeter starts to yield higher stresses over the full edge column perimeter + induced resultant load eccentricity, but being easy to pickup in software these days I tend to up the search radius to grab the free slab edge.

 

[KootK]

Just for your reference, in regard to the smaller sized standard pile caps shown within the CRSI Design Guide for Pile Caps, many of them would have to get significantly larger to meet the 4h requirement, so it seems they're not following it, (e.g. the 4 pile cap is called out as 5'-6" square, but with it's thinnest pile cap of 31", 4h would require a 21'-8" square pile cap, which is obviously unreasonable).

That is interesting, thanks for sharing that tidbit.

Are the CRSI pilecaps that don't meet 4d also taking the full perimeter for punching shear?

 

[JoshPlumSE]

jochav: Does ACI specify the minimum edge distance required beyond the critical perimeter for that perimeter to be effective in two-way shear?

Not that I know of. However, I can share a software issue that I ran into a few times when trying to determine what controls....

So, if you are away from an edge you use one type of punching perimeter and one equation for capacity. Then if you're close to an edge you use a different punching perimeter with a different equation for capacity. Right?

Why not check them BOTH and see which one controls? That's the solution that I believe is being done on the software side.

 

[Celt83]

Thank you, KootK.

I appreciate that; I'll see if there's any wisdom from the software people.

Just for your reference, in regard to the smaller sized standard pile caps shown within the CRSI Design Guide for Pile Caps, many of them would have to get significantly larger to meet the 4h requirement, so it seems they're not following it, (e.g. the 4 pile cap is called out as 5'-6" square, but with it's thinnest pile cap of 31", 4h would require a 21'-8" square pile cap, which is obviously unreasonable).

I believe edge/corner punching is one of the failure modes checked by CRSI, just remember the latest CRSI Pile Cap guide is based on ACI 318-14 they have yet to update for 318-19 and all of the wild reductions in shear that come with it.

Edit: Limit State P3, they also consider combined perimeters for adjacent piles under Limit State P2.

 

[canwesteng]

Well you must always take the punch shear perimeter as a minimum, so bare minimum would be a distance of d+column width. After that it becomes a question of if the possible hit on shear capacity and increased flexural stress from unbalanced moments makes your section worse than the interior case.

 

[JStephen]

"For columns near an edge or corner, the critical perimeter may extend to the edge of the slab."- that's in the commentary, R22.6.4, in ACI 318-14.

 

[jochav52802]

My understanding is there is no limit in ACI for which a free edge does not need to be considered.

In my experience for ACI software tends to follow ACI 318-19 figure R22.6.4.3 (bottom left) and extend the side perimeters in a straight line to the free edge. You usually give the software a radius to search withing for critical sections and the software designs for each complete shear perimeter it can form.

I haven't tested this myself but imagine something like 10*h probably gets you in the ballpark where the complete interior column perimeter starts to yield higher stresses over the full edge column perimeter + induced resultant load eccentricity, but being easy to pickup in software these days I tend to up the search radius to grab the free slab edge.

Thank you, Celt83!

I typically use RISA products; would you please clarify which program you're using that performs the search?

Sounds like I essentially need to run all the possibilities to find the weakest perimeter, eh? I appreciate your help!

 

[jochav52802]

That is interesting, thanks for sharing that tidbit.

Are the CRSI pilecaps that don't meet 4d also taking the full perimeter for punching shear?

You got it.

That is correct, they only appear to check the full circular perimeter, no matter how close they get to the edge.

 

[cliff234]

The critical slab edge dimension beyond which you can consider punching shear on all 4 sides is 4h (per ACI PRC-352.1-11)

 

[jochav52802]

Thank you, cliff234; this is what I was looking for...it seems ACI 318-19 could more explicitly specify that the 4h requirement applies to both openings and free edges.

Based on what you provided, in regard to the below figure from ACI 318-19, Fig. R8.7.7, any thoughts on whether the outermost critical section when headed studs are used for punching shear reinforcement becomes the two red lines or blue lines? It seems to me that it would become the red lines since the (2) green lines I circled in magenta won't have 4h of edge distance. If it's the red lines, I wonder if that will now trigger any changes in the headed stud distribution that ACI 318-19 specifies; looks to me that maybe diagonal studs would now be required perpendicular to the lines I circled in magenta. Thanks for your thoughts!

 

[Celt83]

To me the possible failure planes are the original dashed green, the red and the blue.

In my opinion it is a bit analogous to block shear here where the total blue perimeter has more bo than the red so the red is likely to have a higher peak stress.

I don’t believe this would require diagonal stud rails as the red perimeter just renders the left-right rails ineffective.

All three perimeters should be checked.

 

[jochav52802]

Thank you, Celt83.

Punching shear failed at d/2 from the column face, so I've got to provided shear reinforcement per ACI 318-19 out to the point where the concrete is able to carry the reduced shear stress. My plan is to design stud rails for the original outer green dashed critical section and then also check the parallel green and red lines to determine if I need to shift out the free edge/s at all to get enough moment strength.

On a side note, I've got an assortment of piles under my pile cap close to where fixed-base column loads are applied. Any thoughts on whether I need to consider them somehow on restraining the punching shear load, or does the punching shear limit state always just need to be checked irrespective of how that section may be supported below?

Thanks again!