Eccentric footings and combined footings 5

[pattontom]

In our office, colleagues often use load reduction to design footings with the columns directly on the exterior portion of the lot (zero clearance) to avoid oversize footings. This often results in underdesigned exterior footings. Now I need to design one that needs to use the full commercial live load and dead load for 4 storeys. The footings that come out of SAFE are about 50% bigger. The lot has 12 meters distance end to end and 3 columns spanning them (or 6 meters beam span). I'm thinking of using combined footings for the 3 columns in a row. But haven't seen much book references that treat such for most combined footings often use 2 columns only. Do you recommend that I use 50% bigger spread footings for the eccentric footing or one combined footing for the 3 columns? Which would have better seismic resistance, for those with experience of familiar with them? I understand the combined footing is more sensitive to variable changes in loading affecting the shears and moments in the span. Thank you.

 

[dhengr]

Pattontom:
Are you a real Structural Engineer? Your questions and lack of basic understanding of structural principles and concepts brings this question into play. This sounds like a pretty big project to be cutting your teeth on, and with some typical structural complications too. And, now you say there are two other engineers helping you think they way you are thinking. Given the types of questions you are asking it would seem wise, to me, that you find a local mentor, who has been through this type of construction and design before, who knows the local codes and methods of doing things, but also understands basic Structural Engineering methods, and who could sit with all of you and talk some of this stuff over. You could be looking at the same set of plans, seeing each others sketches and facial expressions as you discussed these things, back and forth. He can generally see when you aren’t understanding what he is trying to explain, and try a different approach. Several real smart guys here aren’t having much luck changing your thinking. That’s tougher to do from a distance and with someone who we have had no experience with. You can learn an awful lot from these older fellows who has been in the structural/construction wars for a longer time, and have seen some of these problems before. Contrary to your protestations, your country does not have different laws of physics and gravity than our countries do, and the mechanics by which we design structures don’t really change from country to country either. There might be minor variations in code criteria, but even those are not appreciably different. No one, not even in the Philippines, can put a large column load on the edge of a spread footing without providing some means of accounting for this eccentricity; either with a combined footing, or tie beams as one of your sketches shows, or by moment capacity in the columns as BA suggested.

 

[BAretired]

pattontom in black, BA in red.

paddingtongreen and BA, in the soil bearing diagram, all axial loads are the same. In diagram in the middle with eccentric load but normal size footing, the column at edge can produce tension in 90% of bottom. If the column at edge is hinged to the footing, it is an eccentric load. It produces variable compression on the soil near the edge, but cannot produce tension because soil cannot take tension. So it produces a triangular block of compression near the edge. But in the rightmost picture with oversize footing and same axial load with respect to left diagram, the bending and shear and sheer weight of the footing can compensate for the lifting and make it stable. This is simply not true. This is what SAFE shows. I don't think so. BA, what do you mean the column has to be designed for bending, it doesn't bend because the 2nd floor has beams that directly connect the left and right side of the lot and all columns connected continuously in upper floors (starting with second) and the floor diaphragms makes it act as one unit. Draw a Shear Force and Bending Moment diagram for the footing. The moment taken by the column is the footing load multiplied by the eccentricity to the center of column. This is the setup for millions of homes in our country. What do you think is wrong? I think what is wrong is your understanding of the situation. Anyway. Is it specifically mentioned in the ACI that the column needs to be located inside the kern? No, it is not nor does it need to be. Next year our structural code which is copied from the USA will be updated. Maybe I can convince the committee to include the kern thing and make all new building permit obey it? Don't even try because that would be wrong. For that to be true, we need to master the art of combined footings which is quite rare here. For example, the setup I'm doing have not been done by more than 80% of engineers here and they haven't handled such thing before except mostly mat and piles foundation. No comment on that as I do not know your local practice.

BA. What do you mean of North south wind on the shearwall? I don't have any shearwall, the left and right walls are 6 inches Concrete Hollow Block filled with cement. An earlier comment of yours stated that the left and right walls are full 6" and I assumed they were shear walls for lateral forces acting in the north-south direction. I am assuming North at the top of your sketch. The back has wall that is 1.5 meters beyond the columns (cantilevered), the front is open as the place would be lease to banks in the future. Also remember the 3 strips of combined footings will be connected to the upper floors solidly. So how would smaller width of the combined footings in each of them overturn the building longitudinally (which is longer than the transverse side) when they are connected as a unit? The shear walls must be effectively carried down to footing level or some other bracing supplied to prevent overturning of the combined footings about an East-West axis.

Thanks guys. So I'll use combined footings. Anyway. The rear and back of the ground floor is open and the second floor front and back cantilever 1.5 meters forward and back. Only the left and right side has full 6" thick wall. My tranverse side is weak. I rely solely on the column-beam joints to give strength to the open front and back. Anyone can comment about open story? Are you confident that joints moment frames are enough? I can't use shear walls because the front and back has to be opened. Any suggestions? Thanks.

BA

 

[pattontom]

Oh.. he was referring to the biaxial bending moments of the columns.. of course, it's taken care of. We mostly use staad and etabs in getting the moments of the structures. We don't do manual calculations anymore here (so many structural engrs. even forgot the formulas). After we got the moments in the software. We enter them in either PCA columns, PCA beams, SAFE or other software to get the reinforcements required for shears and moments.

It's just that we commonly use zero clearance eccentric footings even at corner of building. Our local codes don't disallow it. Maybe it's time we really update our codes. The reasons the building don't fail even with fully eccentric footing is because the load combinations we used for the moments and shear is high at say 1.4DL + 1.7 LL or 1.2DL + 0.5 LL + 1 EQ. Now with actual service loadings that is way down, the actual bearing pressures of the soil actually applied is very low. This is how we get away with zero clearance eccentric loading.

 

[pattontom]

BA,

We only used shear walls for buildings above 5 storey. We never use shear walls for 4 storey and below. The walls in the pictures are all concrete hollow block that only needs wall footing and not shear wall foundation. The building seismic resisting members are all special moment frames in both the transverse and longitudinal directions. For my layout and 4 storey. Do you think this is sufficient? In your country up to what storey can you use pure moment resisting frames as the complete lateral resisting system?

 

[hokie66]

pattontom,
He was not referring to biaxial moments, only the column moment which resists the footing moment.

One more try here. Are you familiar with the principle of superposition? Probably not, if you just trust computer programs to do your job for you. But on the off chance that superposition rings a bell, try this: Take your big footing and place it on the ground. That gives you a uniform soil bearing pressure upwards, resisting just the mass of the footing. Now supermpose a force (your column load) at the edge of the footing. What happens? If there is just an applied force, the force goes straight down through the edge of the footing into the soil below. If you want to distribute some of that force to the rest of the footing, you have to provide a moment connection of the column to the footing. Do you ever design cantilevered retaining walls? Do you have a moment connection between the wall and the footing? Of course you do. I don't know why you fail to see that the same principle is in effect here.

Back to your computer programs...if you just believe everything that is output, you are just a technician, not an engineer. I imagine there are a lot of competent engineers in the Philippines, so won't take your comments about standard practice there as accurate.

As to shear walls, in Australia we commonly use shear walls wherever they are available, even in single storey structures.

 

[pattontom]

Of course it's handled. The column bars are connected continous to the footings. I don't use dowels. For the original oversized exterior footing (the middle left side) with column at the edge most portion of the footing. The specs are the following:

Service Load: 730 Kn
Soil bearing capacity: 150 kpa
Length of footing: 4.6 meters Fc' 20.8 mpa
Width of footing: 3.0 meters
thickness: 0.650 meters
Bottom bars along length: 17 pcs of 20mm bars grade 60
Bottom bars along width: 11 pcs of 20 mm bars grade 60
Top bars along length: 17 pcs of 20mm bars grade 60
Top bars along width: 11 pcs of 20 mm bars grade 60

With this. There is uniform soil pressure at the bottom of the 4.6x3 meter footing with
service load of only 730 kn even a zero clearance completely eccentric column at edge of footing. Work of genius? Lol.

But I'll try combined footing because the sizes of this and other footings differ by 50% that I fear differential settlement can occur. So I'll tell the other senior engineers to start working on the combined footing BA suggested. I admit I'm just a junior member though. Thanks for the ideas and clarifications and tips.

 

[hokie66]

The soil pressure would not be uniform, but let's say it is. Using your 730 kN service load, that would give a soil pressure on your size footing of 53 kPa. This would give you a service load bending moment at the face of the column (assuming a 500 column) of 730 kNm. This bending then has to be taken by the column as well as the footing. Can a 500 x 500 column resist that much moment? No, it can't.

 

[pattontom]

Those numbers come from the SAFE program and you are right that have to manually confirm everything the program says. Hmm.. how would combined footing design help then in making the edge column resist the moment? Remember in the combined footing version the column is still at the edge most location. A combined footing only distribute the soil pressure to the whole combined footing encompassing other columns.

 

[hokie66]

In a combined footing, the columns just deliver forces to the top of the footing. The footing then distributes these forces to the soil by bending. The model assumption would be that the columns are pinned to the footing, although if required to resist lateral forces, they could be moment connected. In the pinned base model, the footing would have zero bending at the exterior columns.

But if you have an isolated, offset pad, then the column has to deliver both the force and a moment which balances the eccentricity of the soil loading. A pinned base doesn't work.

I suggest that you contact the SAFE technical staff for assistance in interpreting the program output.

 

[pattontom]

Using a 0.5x0.5 meter column. Do you know the theoretical axial load limit, like maybe 100-200 KN only, where a zero clearance eccentric column on edge of footing can work (can take the bending moment of any size of footing)? Maybe 730 KN would be so heavy.. so there must be a threshold where it can be done. Any idea what it is? Maybe the reasons millions of buildings in the Philippines use columns at edge of footing is because the loading is so low that it can theoretically take it.

 

[hokie66]

Well, take a 150 kN column load. With 150 kPa allowable pressure, you would need 1.0 m^2. So you could use a 500 x 2000 footing. No eccentricity. If there is any eccentricity, the soil pressure is not uniform, and the column has some bending moment.

 

[pattontom]

I'm asking of eccentric footing where the column is at edge of footing. Because if the combined footings very expensive and client can't afford it. Then I'd advice him to just design building for 2 storey with metal roof. But since his columns need to be at edge of footing with maximum eccentricity. Then the design has to be isolated spread footing with maximum eccentricity. I wonder what is the maximum axial load theoretically possible for this any size of column. I use Etabs and Safes all the time that I forgot how to compute manually. And I want to verify it with manual computations first. So please bear with me this last questions. Thanks.

 

[paddingtongreen]

I believe you said the system is used on one and two story buildings. I suspect that the weight of the bigger foundation is moving the center of load away from the column, this will bring small relief. I also suspect that the column is taking moment and that these effects plus eating into the safety factors is what is holding up the buildings. With the weight of four floors, the weight of the foundation is less significant, that is, it will move the center of load less than with just two floors.

A problem for those of us who are professional engineers is that we cannot live with something we cannot cover with calculations that we understand.

If you draw a free body diagram of your right hand drawing, take moment6s about the point where the column hits the foundation, you will find a resultant moment that can only go into the column. If it is hinged, the middle diagram is what you get, and the top of the footing needs to be reinforced.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[hokie66]

Eccentricity always has consequences. No exceptions. Maybe you think I am eccentric. Maybe I am.

Complication in the footings is no reason for reducing the scope of this project. You already had tie beams, just make them big enough to work as rectifying strap beams.

No offence, but using computer aids without any knowledge of how or why the results are generated is a much more dangerous thing than eccentric footings.

 

[BAretired]

How far below top of Ground Floor Slab is the underside of footing? Section 4/S-5 "Elevation of Tied Column" is not to scale, but seems to indicate a substantial difference in elevation between slab and footing. In other words, how much room do you have for grade beams?

So far, we have been talking about using combined footings in the transverse direction. But there are lateral forces in the north-south direction as well which must be transferred to the foundation. Foundation design must take into account all forces.

The three pad footings on Grid X2 can be centered under the columns with no difficulty. Perhaps a strip footing could be used on Grids X1 and X2 with a grade beam over it to distribute the column loads. In this way, the eccentricity would be small and the column may be adequate to handle the resulting bending moment.

Alternatively, msquare48's idea of a mat foundation could be explored.



BA

 

[ron9876]

Plenty of experienced engineers have responded here to the technical aspects of the original question so I will stay out of that conversation.

Pattontom my only comment is that you have stated that you do not have a mentor yet you are using a very powerful program like SAFE. To use that type of program without knowledge and experience in performing the design by manual calculations is a major concern. This is exactly what worries some of us gray hairs that have seen the next generation of structural engineers come along that simply trust the output from a program. I agree with the recommendations above that you find a local engineer that has the knowledge and experience to design the structure. Good luck.

 

[pattontom]

Ron,

Other engineers handling this and the calculations. I'm just one of them and just offering suggestions esp when eccentric loadings are normally done by anyone here.

BA,

The grade beam or tie beam are actually just below the slabs where the beam supports the slab.

About forces in the lateral direction. I thought the larger width of the combined horizontal footings can take care of that. Remember in isolated spread footing, it can take tranverse and longitudinal forces well.

When you mentioned "The three pad footings on Grid X2 can be centered under the columns with no difficulty. Perhaps a strip footing could be used on Grids X1 and X2 with a grade beam over it to distribute the column loads. In this way, the eccentricity would be small and the column may be adequate to handle the resulting bending moment." Are you talking about separate strip footing and grade beam in X1 and X2 in addition to the 3 combined horizontal footings or separate?

Mat foundation may be overkill. Anyway making the width of the 3 horizontal combined footings can take care of the lateral forces in the north-south direction? No? I'm still waiting for the engineer in charge of it to do the calculations. I can only offer suggestions.

Hokie66 you said "The soil pressure would not be uniform, but let's say it is. Using your 730 kN service load, that would give a soil pressure on your size footing of 53 kPa. This would give you a service load bending moment at the face of the column (assuming a 500 column) of 730 kNm. This bending then has to be taken by the column as well as the footing. Can a 500 x 500 column resist that much moment? No, it can't."

How did you get the value of the 730 KNm moment?

 

[hokie66]

Your column load was stated to be 730 kN. Using your assumption that the soil pressure is uniform, then the upward soil pressure is also 730 kN (ignoring little things like the mass of the footing). So you have a force upwards of 730 kN, centred at a distance of 1.0 metre from the face of the column. 730 kN x 1 m = 730 kNm. Simplified and not correct, as the soil pressure would vary, but illustrative of the problem you have in developing such a large moment without a big column.

 

[pattontom]

To get design soil pressure, formula is qu=Pu/A
The service load is 730kn
Dead load is 500 kn. Live load is 230 kn.
So Pu=1.2DL + 1.6 LL = 1.2(500) + 1.6 (230) = 600 + 368 = 968 Kn
qu = 968 Kn/(footing area)^2
= 968 Kn/13.8^2 sq.m
= 70. 14 kn/m^2 or kpa

Now. Moment = WuL^2/2
L=(4.6-0.5)/2 = 2.05
wu = qu (1 meter ) = 70 kn/m

Mu= WuL^2/2= 70kn/m (2.05)^2/ 2 = 147 kn.m

so moment of footing is only 147 kn.m and not the 730 kn you mentioned.

 

[BAretired]

Are you talking about separate strip footing and grade beam in X1 and X2 in addition to the 3 combined horizontal footings or separate?

I was trying to suggest an alternative foundation system consisting of a strip footing on Grids X1 and X3 (I typed X2 in error) and three square footings, each centered under a column on Grid X2. The advantage is that the strip footing would have a small width, hence a small eccentricity which, perhaps could be carried by bending in the exterior columns.

A second possibility is to provide a strip footing all around the entire perimeter plus one single square footing under the central column on Grid X2.

The strip footing would have to be stiff enough to distribute the column loads over the length of footing. I don't know whether this would be practical or not but it is worth exploring.

BA