Point Load Near Edge of Mat Foundation

[P1ENG]

Once again, I have a concrete question. I have a 71.6 kip LRFD load coming down an 8"x8" tube column. There is an existing 18' hexagon concrete pad (3' thick) on the site. The new construction hits the existing pad on the edge as shown below. The 8" tube is flush with the edge of the concrete. How can I check the existing hexagon pad (failure methods? shear? blowout?,etc.)? I can't traverse the ACI code so well, and I don't ever find what I need with any confidence. Seems like actual analysis methods come from PCA whenever I need to concrete calculations. Please point me in the right direction. Maybe it is as simple as assuming a strip of concrete with a width of 25.5" (8.75"+8"+8.75") and just doing a SCP shear check of a beam?

First, I am assuming unreinforced concrete since we know little about the existing concrete.

If I can't get the existing pad to work, I guess I will make them cut the hexagon out such that I can get a proper square footing in there? It is a large mass of concrete, so I would like to use it if I can instead of going through the labor of cutting it out. I also thought about doweling the existing pad and pouring new against it to get the required square footing. Please hit me with some suggestions.

Juston Fluckey, SE, PE, AWS CWI
Engineering Consultant

 

[WARose]

How can I check the existing hexagon pad (failure methods? shear? blowout?,etc.)?

Looks pretty straight forward to me: a hand check on the shear to start out with on the punching (i.e. 2-way shear). (I doubt very seriously one-way would control (it seldom does).) I can't imagine a 36" foundation having a problem with that load as far as shear goes.

The next thing is possibly a FEA model with the foundation spring supported to get the moments. Hopefully you have some geotechnical info to get the necessary spring constants. Maybe a hand calculation (with it on a rigid surface) to be sure you are ok as far as overturning (and you have full bearing)....just to supplement the FEA model.

First, I am assuming unreinforced concrete since we know little about the existing concrete

So I assume you have no drawings? Not good. If you can, might be best (and cheaper) to just have a independent foundation than a bunch of testing to find out what the re-bar is in the exiting mat.

And keep in mind: ACI code prohibits plain concrete foundations in Seismic Design Category D, E, or F. (Unless we are talking residential in some circumstances.) Me personally, I'm not a fan of plain concrete period unless we are talking something like a slab on grade. (I.e. where it can crack like crazy and there won't be any catastrophic issues.)

 

[BAretired]

Shear would worry me because the column is so close to the edge. I think the failure surface would start at the inside edge of the column and slope down toward the exterior face. It may not even engage the full 36" depth of the existing slab.

For bending, it seems likely that the existing slab is unreinforced.

Doweling into the existing pad would be one option, but it might be more economical to place a new rectangular footing (underpinning) underneath the existing pad with the shorter dimension running perpendicular to the existing concrete face.

BA

 

[P1ENG]

WAR: Thank you. FEA model, I'm sure, is outside of my recoverable fee. My client is providing a tower to sit on top of someone else's foundation. I don't think my client wants me to spend too much time getting somebody else's unknown foundation to work. Now if it is straight forward and a simple check, then sure. Also, this is SDC C, so plain concrete should be ok. Again, I am assuming plain concrete for checks. I would expect there to be some steel in that mat.

BA: Thank you too. My concern was with its proximity to the edge as well. You're thinking the failure plane will start roughly 8" from the edge and angle back toward the edge at an angle of 35 degrees from the vertical, such that I only engage ~11.4" of the concrete depth instead of the full 36" depth? I don't understand your comment - "with the shorter dimension running perpendicular to the existing face." Can you clarify or maybe provide a quick sketch? I was also told to assume crappy soil (2000 psf), so my required footing size when looking at service loads is 5'-9" square. With my footing square, where does the "shorter dimension" come into play, or should I not make the footing square?

Juston Fluckey, SE, PE, AWS CWI
Engineering Consultant

 

[WARose]

[blue](BAretired)[/blue]

Shear would worry me because the column is so close to the edge. I think the failure surface would start at the inside edge of the column and slope down toward the exterior face. It may not even engage the full 36" depth of the existing slab.

Do you feel that way because it (may be) unreinforced? If so, I can understand that.

 

[BAretired]

P1Eng,
You could use a square footing. I prefer a rectangular footing in order to reduce the width of excavation under the existing pad. If P = 71.6 kips and bearing is 2000psf, you need an area of 35.8 S.F. Could use 6' x 6' but I'm suggesting 9' x 4' as shown by the dashed red lines in the attached file. It's really up to you to select the most economical proportions.

BA

 

[BAretired]

Do you feel that way because it (may be) unreinforced? If so, I can understand that.

Pretty much although even if the pad is reinforced, the reinforcement may not extend far enough past the potential failure plane to prevent it from forming. It is a judgment call and that is my judgment.

BA

 

[BAretired]

Upon further thought, I am wondering if the same failure plane could occur with or without the proposed new footing. I'm beginning to think it could. It may be advisable to drill and epoxy some reinforcement into the existing pad to prevent the formation of that failure plane.

BA

 

[P1ENG]

BA: That was my thought as well. The new concrete next to the mat isn't going to stop the mat from doing what it will do. There needs to be some composition. Once I dowel it though, there is no point in underpinning. Do you agree?

Here is my sketch. I laid my square footing centered on the column and trimmed it where it crossed the mat. Dowels were calculated to transfer the new concrete's share of soil bearing. Other reinforcement will be included, but just the dowels are shown.
*EDIT* My failure was concrete breakout, so I don't need that many dowels. I will actually be using 12" o.c. in both directions for a total of (15) dowels instead of the (32) shown.

Juston Fluckey, SE, PE, AWS CWI
Engineering Consultant

 

[BAretired]

P1ENG,
Your concept looks good to me. You might want to put some nominal steel in the other direction.
I agree that there is no need to underpin if you extend the existing pad as you have indicated.

BA