Isolated Foundations Design

Foundations are a fundamental part of any structure, as they are responsible for correctly transferring loads to the ground. Within the wide variety of foundations, isolated footings are the most commonly used for small and medium-sized structures due to their simple design, ease of construction, and high versatility.

The operating principle of foundations is quite intuitive: distribute the load over a larger area so that the soil receives lower pressure and does not fail. This is why multiple civilizations thousands of years ago employed rudimentary shallow foundation systems; even today, many home projects are executed by intuition without calculations using isolated footings. Do we truly understand how these foundations work? Are they correctly designed? Could they be optimized? In this article, you will find the answers to these questions and much more.

Pressure distribution in isolated foundations

To correctly design an isolated footing, it is fundamental to understand how contact pressures are distributed across the bearing surface. This depends on multiple factors, such as the position and components of the load, the soil type, and the stiffness of the foundation.

Load cases without eccentricity

The simplest load case corresponds to a completely vertical load, located at the center of the foundation. In this case, the type of soil on which the foundation is built is the main factor influencing the pressure distribution.

Sandy soils have very little cohesion, considered null in many cases, so the friction generated at the edges of the foundation when it is vertically loaded is practically non-existent. Thus, if we consider that the foundation is infinitely rigid compared to the soil, the pressure distribution is uniform across the bearing surface. If we consider that the foundation is not as rigid, then the force from the column is not uniformly distributed across the bearing surface; instead, there is a concentration of pressures in the vicinity of the column.

On the other hand, clayey soils have high cohesion, which generates significant friction at the edges of the foundation when it is vertically loaded. In this way, if we consider that the foundation is infinitely rigid compared to the soil, there would be a concentration of pressures at the edges of the foundation. In the case of a flexible footing, the effect of pressure concentration at the column tends to be neutralized by the concentration of pressures at the edges. The following image schematically illustrates this behavior.

However, these details in the non-uniform distribution of pressures are usually neglected in the design of isolated foundations. It is widely accepted to consider that the pressure distribution in the case without eccentricity is uniform, determined simply as $\sigma = \frac{P}{A}$ where $P$ is the total load and $A$ is the area of the foundation.

Load cases with eccentricity

The panorama changes completely when we face loads with eccentricity, cases where assuming a uniform pressure distribution is not only incorrect but also potentially very dangerous. Eccentricity in a foundation can essentially be due to two reasons: (i) vertical load with an eccentric position on the footing, or (ii) loads with lateral components (shear or moment).

Eccentricity not only generates higher pressures in some areas of the bearing surface but can also cause footing uplift. Soil has a certain limit for the compression that the footing can transmit to it (allowable stresses), but in no case can it transmit tension.

Winkler Method

In 1867, Emil Winkler revolutionized the way we understand soil behavior by proposing that it could be analyzed as a dense set of independent springs. In this way, the soil is discretized, and each point deforms according to the pressure it receives.

Idealized linear model

In principle, we can understand these soil springs simply through linear Hooke’s Law. That is, each spring presents a displacement that is directly proportional to the pressure exerted on it, with the proportionality ratio being a factor $k_s$ that we call the "Modulus of Subgrade Reaction." Mathematically, this is expressed as:

This idealization works for many analysis cases—specifically, for all those in which the soil remains within the range of linear behavior. However, soil has a great variety of non-linear behaviors, making it necessary to include some additional considerations.

Non-linear models

Among the non-linearities of soil behavior, the most important to take into consideration is that the soil is incapable of generating tension on our foundations. When a foundation loads downward, the soil responds with upward pressure, but when a foundation tries to lift, the soil does not pull it downward. Thus, the main characteristic that must be included in the analysis is a lower bound for the pressures in the springs, as they only allow compression. In Foundaxis, this is implicitly considered in all analyses, but in other softwares, it may be necessary to manually activate the non-linear analysis option and assign this restriction to the springs.

Additionally, there are other characteristics of the soil that should be taken into consideration to achieve an accurate analysis of the deformations generated within it. For example, soil does not have infinite strength, so linear behavior has a limit after which the soil fails. Similarly, depending on the properties of the soil, there may be more complex deformation phenomena, such as soil grain rearrangement or water flows, which are studied with more complex numerical models.

One-way eccentricity

If there is eccentricity of the load in only one direction, the problem is relatively simple, as it can be studied in 2D. In these cases, it is reasonable to assume a linear distribution of soil pressures, and it is enough to study the equilibrium of forces to exactly determine the pressure distribution across the bearing surface. In principle, it is the same equilibrium of forces and moments that determines whether it is a case of uplift or full contact.

In practice, the case of uniaxial eccentricity is almost an academic idealization. The vast majority of columns in a building are subject to bending moments in both main directions due to wind loads, earthquakes, or simply the continuity of structural frames. This leads us to biaxial eccentricity analysis.

Biaxial eccentricity

In the case of biaxial eccentricity, the problem can no longer be analyzed in 2D, as the pressure distribution corresponds to a plane inclined in two directions. The mathematical challenge arises when the eccentricity is so large that footing uplift occurs at one or more corners.

This is where manual calculation becomes a nightmare. When there is uplift, the shape of the effective contact area stops being a full rectangle; it can become a triangle, an irregular trapezoid, or even a pentagon. For each corner, 4 cases of biaxial uplift can occur, as shown in the following figure:

Determining the exact position of the neutral axis (the line where pressure is zero) requires solving complex non-linear equations or using iterative methods. Classical solutions based on coefficient tables or charts (such as Teng or Meyerhof tables) are tedious and prone to interpretation errors.

An incorrect design that ignores biaxial uplift or assumes an erroneous simplification can drastically underestimate the maximum pressure, jeopardizing the stability of the structure. On the other hand, an accurate analysis allows for optimizing the footing dimensions, seeking the right balance between safety and material economy.

Sizing of isolated foundations

In practice, the sizing of foundations is essentially based on trial and error. Although estimates can be obtained with approximations for pressure distribution, it is ultimately through modeling and analysis that the design dimensions are determined, which will hardly be optimal. To accept the design of a given foundation, different performance limit states must be verified.

Soil bearing capacity

Soil bearing capacity corresponds to the maximum pressure it can resist before experiencing a sudden shear-type failure, in which there is an abrupt displacement between adjacent sections of soil. There is no single way to estimate the bearing capacity of a soil, but among all the studies conducted on the subject, it has been concluded that cohesive and non-cohesive soils must be treated differently:

• Analytical Models: For soils with cohesion and friction, shear failure theory is used. The general bearing capacity equation (perfected by Meyerhof, Vesic, or Brinch-Hansen) considers factors of shape, depth, and load inclination, following this style:

  $$q_u = c N_c s_c d_c + q N_q s_q d_q + \frac{1}{2} \gamma B N_\gamma s_\gamma d_\gamma$$

• Empirical Approaches: In sands, it is very common for bearing capacity to be estimated directly through correlations with in-situ tests, such as the SPT (Standard Penetration Test). Methods like those of Meyerhof or Peck relate the number of blows ($N_{60}$) to the allowable pressure to limit settlement to tolerable values (usually 1 inch).

All this information must be consolidated in the Geotechnical Report (IMS). This document provides us with the design guidelines. Although this is generally the work of the geotechnical engineer, there are some key points to which the structural engineer must pay special attention when receiving this information and using it in their design:

1. Validation of hypotheses: The geotechnician calculates $\sigma_{adm}$ based on an estimated embedment depth ($D_f$) and footing width ($B$). If our final design deviates significantly from those dimensions, the actual capacity could be different from that in the report.

2. Presence of water: The water table can significantly reduce bearing capacity due to the effect of buoyancy and the reduction of the effective unit weight of the soil.

Additionally, the methodology used for foundation design is based on maximum allowable pressures for the ground, which must be lower than the bearing capacity. Allowable pressures are obtained by dividing the bearing capacity by a factor of safety $FS$. For static loads, it is common to use $FS \geq 3.0$, while for dynamic loads, allowable pressures are approximately 33% higher. This is mainly due to the duration of the loads on the structure; dynamic loads include earthquakes, for example, which can easily last only a few seconds. Thus, for sporadic loads, the soil is allowed to experience higher pressures.

Percentage of contact area

Verifying the percentage of contact area in a foundation is a way to check the stability against overturning. This percentage corresponds to the fraction of the soil-foundation interface that is in compression. Different design regulations indicate different values for this performance parameter; however, a reasonable value is to ensure that at least 80% of the bearing surface is in compression for all service load combinations. Service loads correspond to those to which the structure will effectively be subjected day to day, without load factors.

Settlements

Even if the soil is capable of resisting pressures without failing by shear, it will always experience some deformation under load. In engineering, we call this vertical sinking "settlement." The designer's challenge is to predict not only how much the footing will sink, but in what time and in what manner it will do so.

Mechanisms of settlement

Not all soils respond at the same speed. The nature of the ground determines the "rhythm" of the settlement:

• Immediate or Elastic Settlement ($S_i$): Occurs almost immediately after applying the load. It is predominant in granular soils (sands and gravels) and in dry or partially saturated cohesive soils. It is the type of movement we see during the construction phase.

• Consolidation Settlement ($S_c$): It is a time-deferred process, typical of saturated clays. When loading the soil, the water trapped in the pores is slowly expelled due to the low permeability of the clay. This process can last months or even decades, causing the building to continue sinking long after the work has been finished.

Various soil improvement techniques exist designed to 'force' consolidation settlements before starting the work. A common method consists of installing vertical drains (such as wicks or perforated pipes) combined with a preloading of the ground. By applying a controlled weight over the area, the water trapped in the soil pores is rapidly expelled through these drains, accelerating a process that would naturally take years. Thus, when the final structure is built, the soil has already experienced most of its deformation, minimizing unforeseen movements in the future.

Ways to measure settlement

For the structure, it is not so important how much it sinks in total, but how evenly it does so:

• Absolute Settlement: Is the total sinking of an individual footing. If the entire building settles uniformly by 5 cm, the structure is unlikely to suffer internal damage, although we could have problems with service connections (gas or water pipes) or access levels.

• Differential Settlement ($\Delta S$): Occurs when one footing settles more than the one next to it. This generates angular distortions in the beams and slabs, causing cracks in walls, misalignment of door frames, and, in severe cases, structural failures due to unforeseen stresses.

• Key Fact: Most standards limit the differential settlement between adjacent columns to very strict values (such as $L/500$ or $L/300$ of the span between them) to protect the integrity of the finishes and the structure.

This is where isolated foundation design becomes an art of balance. A very rigid structure can "bridge" small differences in settlement, but at the cost of generating large internal stresses. Conversely, a flexible structure accommodates the ground better but is more prone to visible cracks.

Structural design of isolated foundations

There are design regulations that, for certain low-load cases, allow foundations to be simply a block of concrete (e.g., plain concrete in chapter 13 of ACI 318-25). However, in the vast majority of cases, foundations require steel reinforcement.

Bending

In isolated foundations, it is most common for the bottom face of the foundation to be subjected to tension due to bending, regardless of whether the load is eccentric or not. This is because concrete is very strong in compression but very weak in tension. In cases of high eccentricity, low height, or very large foundations, it may also be necessary to specify steel reinforcement on the top face.

The specification of steel reinforcement in foundations must be done with special care, as they are not only crucial elements for the structure that are subjected to the harshest environmental conditions, but they are also hidden from any visual inspection that helps identify structural pathologies. Thus, a very good practice is to specify a larger concrete cover for the foundation reinforcement, which can be up to 5-7 cm of clear cover. This is with the objective of protecting the bars from moisture and corrosion.

Punching shear

When an isolated foundation is large enough, it can experience a punching shear failure in the same way as a slab. The loads coming from the pedestal are not immediately distributed across the entire area of the foundation; instead, they do so gradually, inducing shear stresses within the foundation.

A shear-resistant perimeter is defined to verify punching shear failure. If a foundation does not have special reinforcement for punching, the resistant perimeter is simply a rectangle around the pedestal. A foundation can also have specific reinforcement for punching, which usually consists of a pair of "embedded beams" in the foundation; in this way, the shear-resistant perimeter grows considerably, taking an octagonal shape as shown in the figure:

It should be noted that in the figure both cases correspond to pedestals centered on their foundations, which is not always the case. Depending on the proximity to the edges, there may be edge pedestals (3-sided perimeter) and even corner pedestals (2-sided perimeter). In practice, it is not common to specify punching reinforcement. If a foundation is failing by punching shear, the simplest and most widely used solution is simply to increase the height of the foundation.

One-way shear

Just as punching shear evaluates two-way shear resistance with a shear-resistant polygon, the one-way shear verification evaluates the resistance to shearing an entire strip parallel to one of the sides of the foundation, considering a straight line as the resistant "perimeter." This verification usually has greater relevance in rectangular foundations with an aspect ratio $B/L \gtrsim 1.5$, where one dimension is considerably larger than the other. The failure lines in both directions are schematically presented in the figure:

Again, in the event that a foundation does not comply with this verification, the simplest and most widely used solution is to increase the height of the foundation and not specify shear reinforcement.

How to design with Foundaxis?

In this article, we have reviewed that the design of isolated foundations is not trivial; there are multiple variables to take into consideration and complex phenomena that it is extremely important to understand. Foundaxis completely automates the design process, optimizes the foundation dimensions to save on material use, calculates the necessary steel reinforcement, and performs the geotechnical analysis, all in a few minutes.

However, Foundaxis does not force you to blindly trust its results. Within the application, there are various panels to review each of the design and performance results of your foundations in detail. Additionally, a fully detailed calculation report is generated in a couple of minutes, which you can download and edit as you wish, the same with quantity take-offs and detail drawings.

Foundaxis does not do the work for you, but it makes your work much faster and more fluid, allowing you to concentrate your energy on the processes that matter most.

 

Discover how isolated foundation design is integrated into the complete Foundaxis workflow:

Foundation Design Software: A Complete Guide for Structural Engineers 2026