Mat foundation design by hand

[bldgdesign]

Anyone have any good references for designing mat foundations by hand. Not combinded footing with two columns, but mats that support equipment with 8 to 10 columns or mat foundations that support 10 to 12 pieces of equipment such as pumps, horizontal and vertical fuel tanks (10' diameter), etc.

I am trying to justify buying software to do the design of 10 different mat foundations on a project. I have used software to design all mat foundations in the past, but this company doesn't have software.

I would also just like to have the reference.

 

[Lion06]

I have some examples from my undergrad foundations course. I believe there is also a couple examples in the Das foundation engineering text.
Do you really want something to go by or do you not so that there is a reason to buy the software?

 

[bldgdesign]

I really want something. I didn't have an undergrad foundation course. I have done a few hand calcs for transfromers and other equipment foundations but always used pca-mats for multiple equipment. If you can reference something for me that would be great.

 

[PanamaStrEng]

Why not use a general FE program with foundation springs? SAP or RAM can do it. That is exactly what PCA MATS does.
The hand calculations can get nasty real quick, and in the end you'd end up assuming a lot of things.

 

[edwin17th17]

i wonder how you deal with FE program when it comes to putting loads instead of pedestal.. which will cause the surrounding element to be too stressed and will give you unrealistic rebar using the stress of that element..

what i do is neglect the element right next to the point load and consider only the next element for rebar design..

any opinion..

 

[Lion06]

Hand calcs aren't bad as long as you make the assumption that the foundation is rigid, it's just a lot of book-keeping. You basically treat it like a cross-section with all those various column loads on it. You find the resultant of those loads (and it will likely have an eccentricity in both directions). From that, take P/A +- P(ey)y/Ix +- P(ex)x/Iy in order to get the bearing pressures at each column. From that, you can treat it like a series of individual beams with the load being the soil pressure and the supports being the columns. Pick the worst case beam in each direction and design accordingly.
Don't forget to check for punching shear.

 

[PanamaStrEng]

edwin17th17: I guess it depends on the size of your mesh. If it is very fine, maybe you could spread the load in more that one joint, or even apply it as a shell pressure (if there are no moments; if there are, its easier to use a single joint). Even then, high stress concentrations may ocurr. You cannot ignore them, but you can obtain the total moment in a "design strip" that includes the offending stress concentration and spread it over a wider section whihc is where the reinforcement will be located.

 

[edwin17th17]

"but you can obtain the total moment in a "design strip" that includes the offending stress concentration and spread it over a wider section whihc is where the reinforcement will be located."

i didnt get what you mean by this.. are you talking about creating a strip and doing a hand calculation of shear and moment diagram?

 

[PanamaStrEng]

Most FE programs will produce output for moments in units of F*L / L. You could "integrate" those results into a known width of your choosing in order to place reinforcement. Generally stress concentration take place over a very small width.
Not doing a hand calculation for moment and shear, but using the results to obtain workable values.

 

[rn14]

I have been using SAFE to do bunch of foundations lately. The nice thing about using it rather than say SAP or Risa-3D with comp only springs is that you get bearing pressure also.

If I have a piece of equipment with a larger base plate I will usually apply its load as an area load to try and avoid the large concentrated load.

 

[Lion06]

rn14-

Couldn't you very closely approximate your bearing pressures when using compression only springs by dividing your max spring reaction by the area of footing it support (i.e. max spring reaction/(distance between springs * width of footing))?

 

[rn14]

Definatly. It is just nice to be able to print out a nice colorful contour of the bearing pressure.

 

[edwin17th17]

panama,
but still maximum moment would come from the most stress element that adjacent to the point load..

 

[PEinc]

Foundation Analysis and Design, 1968, by Joseph E. Bowles, pages 223 - 229, has discussion and a design example for mat design.

 

[abusementpark]

I've used PCA Mats and think it is a fine program. If you just need software for this one project, you can at least download the 15-day trial version of PCA mats to get some preliminary designs done.

 

[PanamaStrEng]

The element adjacent to the load will for sure be the one showing the stress concentration and maximum moment. Try and refine your mesh around it. You´ll notice that the area showing high stresses is "short lived". This high stress is also physically not true.
If you tell the software to show you moments, you could add up the maximum moments per unit length of each element, multiplied by the width of each element on an assumed design strip and design for the total moment in the strip.
As you stated, bearing pressure is estimated as the force in the compression only spring divided by its tributary area.