0% found this document useful (0 votes)
283 views7 pages

Practical Formula for Rectangular Footings

This document presents a practical formula for determining the minimum dimensions of a rectangular footing subjected to biaxial bending and a vertical column load. The formula is based on the linear distribution of pressure at the footing-soil interface. It considers the column load, moments, lateral loads, footing weight, and allows the footing to be fully compressed without tension. The formula provides an algebraic solution as an alternative to iterative methods typically used for footing design. It unifies various formulations and can be validated by comparing results to classical methods.

Uploaded by

prisciliano1
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
283 views7 pages

Practical Formula for Rectangular Footings

This document presents a practical formula for determining the minimum dimensions of a rectangular footing subjected to biaxial bending and a vertical column load. The formula is based on the linear distribution of pressure at the footing-soil interface. It considers the column load, moments, lateral loads, footing weight, and allows the footing to be fully compressed without tension. The formula provides an algebraic solution as an alternative to iterative methods typically used for footing design. It unifies various formulations and can be validated by comparing results to classical methods.

Uploaded by

prisciliano1
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 7

Engineering Structures 29 (2007) 1128–1134

www.elsevier.com/locate/engstruct

Practical formula for dimensioning a rectangular footing


H.M. Algin ∗
Department of Civil Engineering, Faculty of Engineering, Harran University, Osmanbey Campus, 63000, Sanliurfa, Turkey

Received 23 March 2006; received in revised form 9 August 2006; accepted 16 August 2006
Available online 28 September 2006

Abstract

The relationship among the flexural formulae and the vertical and rotational equilibrium of forces is correlated analytically to develop a simple
and unique expression to determine the required minimum footing area under full compression. The current practice of estimating the size of
footing in structural design usually employs the conventional iterative process initiated with the educated guess of designers. The presented
formula is a practical algebraical solution as an alternative to the conventional trial and error method for estimating the minimum dimensions of
a rectangular footing subjected to the combination of biaxial bending in both axes and vertical column load. The weight of footing and the lateral
column loads are also included in the formula. It is straightforward in appearance and application. It is based on a linearly distributed surface
pressure regime which disallows the tension to be developed in the footing/elastic-medium interface. It unifies the various formulations common
to the field of structural engineering. Paper explains the validity of the new formula by comparing its results with those produced by classical
methods. A simple design chart is also provided to determine the minimum dimensions of a rigid rectangular footing resting on elastic mass.
c 2006 Elsevier Ltd. All rights reserved.

Keywords: Buildings; Structure; Design; Footings; Foundations

1. Introduction weight of footing, the lateral column loads and the column
moments about both axes. The various combinations of dead,
Conventionally, the design of rectangular footings is live and wind loading may be introduced by a centrally placed
concerned initially with the assumption of suitable footing column.
dimensions to ensure that under serviceability conditions Footings are commonly assumed to act as rigid structures.
excessive ground bearing pressures are not developed [1–5]. This premise leads to the conclusion that the vertical settlement
Once a suitable base size has been calculated the base can be of the elastic medium beneath the base must have a planar
designed for bending, shear and punching shear at ultimate distribution because a rigid foundation remains plane when
load conditions. The presented formula calculates the minimum it settles. A planar distribution of soil pressure follows from
dimensions of square or rectangular rigid footings without a second assumption that the ratio of pressure to settlement
any iteration or initially selected footing contact area that is is constant. This results in an assumption that the pressure
subjected to the combination of biaxial bending and vertical distribution below a rigid footing is to be linear. Neither of these
column load. It is analytically derived for rectangular footing assumptions is strictly valid, but each is generally considered
areas under the full compression and two-way eccentricity. In sufficiently accurate for ordinary problems of design. In fact,
most design cases the tension between the footing and elastic most soils show evidence of some plastic behaviour, footings
mass is not required to develop, so in this case the whole generally have a finite stiffness and the distribution of footing
footing area is of full compression. However, the solution pressure within soil varies with time. The actual distribution
technique presented here may be used for any type of linear of footing pressure within soil depends on the type of soil and
load intensities assumed to be developed under a rectangular
stiffness of the footing and the structure. Since the behaviour of
footing. The formula includes the axial column load, the
foundations involves many uncertainties regarding the action of
the ground and the loading, the soil pressure distributions rather
∗ Tel.: +90 414 344 0020; fax: +90 414 344 0031. than linear are usually not considered in present-day footing
E-mail address: [email protected]. design practices [1–5]. The linear soil pressure distribution

c 2006 Elsevier Ltd. All rights reserved.


0141-0296/$ - see front matter
doi:10.1016/j.engstruct.2006.08.009
H.M. Algin / Engineering Structures 29 (2007) 1128–1134 1129

Notation
B width of footing area
ex eccentricity distance at the x direction
ey eccentricity distance at the y direction
f a = qa maximum soil pressure
h height of footing
Hx lateral load of column about the y axis
Hy lateral load of column about the x axis
L length of footing area
Mx moment of column about the y axis Fig. 1. Linear pressure configuration adopted by Jarquio [6].
My moment of column about the x axis
Pc vertical load of central column
qa maximum load intensity at one of the corner of a
rectangular footing area
qb , qc and qd load intensity at corners of a rectangular
footing area
V volume of pressure configuration regime
x̄ centre of gravity distance at the x direction
ȳ centre of gravity distance at the y direction
γ unit weight of footing or average unit weight of
overburden soil and footing

assumed in this paper is mostly used in typical day-to-day Fig. 2. The general form of a linearly distributed pressure configuration at the
bottom of rectangular footing [12,13].
footing design solutions [1–5].
The complexity of dimensioning a rectangular footing
area with iteration has brought about the need to develop author [12,13]; it can be applied directly to entire linear surface
more general and accurate design configuration. One of the variations that include the solutions previously developed for
analytical formulae for determining biaxially loaded footing those subsets [6–8,10,14]. The similar linear configuration
area was developed by Jarquio [6]. The linear contact pressure is also adopted in this paper to facilitate more general and
configuration adapted by Jarquio [7] was published by Vitone accurate design formula for dimensioning rectangular rigid
and Valsangkar [8]. They presented an arithmetical solution of footing under biaxial bending.
stress distribution for a linear Boussinesq [9] loading case by
applying the method of superposition to specific triangular load 2. Minimum footing area
configuration which was formulated by Gray [10]. The solution
of the linear pressure configuration illustrated in Fig. 1 was The proposed formula is based on the vertical and rotational
adopted by Jarquio [6] for dimensioning the rectangular footing equilibrium of the pressures applied to the upper and lower
area subjected to biaxial bending. horizontal surface of rectangular footing resting on an elastic
Although it is a common practice for many researchers [6, half-space. It is assumed that the footing is structurally rigid
8,11] to equalize one corner of pressure block to zero, in relation to the elastic mass so a planar pressure regime is
Jarquio’s [6] solution is valid only when the load intensity assumed at the footing/elastic medium interface. In most of
value at one of the corners is of zero contact pressure, when the design cases the tension between the footing and elastic
the load intensities at each of the two corners are equal and mass is not required to develop, so in this case whole footing
when the maximum load intensity at the remaining corner is area is of full compression. This condition is made possible
twice the value of the pressures at the two corners (see Fig. 1). by setting one corner of the rectangular footing with a lower
For the plane load intensity variation shown in Fig. 2, the pressure intensity (qd ) and the diagonally opposite corner with
condition of one corner having zero contact pressure does not the maximum allowable elastic medium pressure ( f a = qa )
necessitate that the other two corners are equal to the half of
(see Fig. 2). From the geometry of the pressure configuration
the maximum load intensity at the remaining corner. Corner
shown in Fig. 2, the following equations for the volume (V )
load intensities might have different values whilst satisfying
and the centre of gravity distances at the x and y directions (x̄
the design conditions and providing the minimum contact
and ȳ) of the pressure block can be obtained in terms of the load
area (see Numerical Examples). Generally, setting one corner
intensities at the corners (qa , qb and qc ) and the dimensions of
of pressure to zero is not a sufficient condition to have a
minimum footing area. The general close solution of stress footing contact area (B × L).
distribution in soil mass for linearly distributed surface pressure B L(qb + qc )
resting on an elastic half-space was previously published by the V = (1)
2
1130 H.M. Algin / Engineering Structures 29 (2007) 1128–1134

B(3qb + 4qc − qa ) B Mx + h Hy
= − . (8)
6(qb + qc ) 2 Pc + B Lhγ
Eqs. (6)–(8) are one vertical equilibrium and two rotational
equilibrium equations, one in each orthogonal direction. This
ensures the vertical and rotational equilibrium between the
design load configuration and the reaction pressure block (see
Fig. 3). Eqs. (6)–(8) can be solved to obtain the contact pressure
intensity at each corner of the lower horizontal surface of the
footing. The pressure at any point along the footing boundary
can then be determined by calculating surface functions of
vertical pressure distribution. Solution of Eqs. (6)–(8) in terms
of qa , qb and qc produces following equations.
qa
Fig. 3. Eccentrically loaded rectangular footing. 6L(Mx + h Hy ) + B(6h Hx + 6M y + L Pc ) + B 2 h L 2 γ
= (9)
B2 L2
L(4qb + 3qc − qa ) qb
x̄ = (2) 6L(Mx + h Hy ) + B(−6h Hx − 6M y + L Pc ) + B 2 h L 2 γ
6(qb + qc ) =
B(3qb + 4qc − qa ) B2 L2
ȳ = . (3) (10)
6(qb + qc )
qc
Many foundations must resist not only vertical loads but also
−6L(Mx + h Hy ) + B(6h Hx + 6M y + L Pc ) + B 2 h L 2 γ
moments and lateral loads about both axes. The total moments = .
Mx and M y can be located at the bottom of a concentrically B2 L2
placed column (see Fig. 3). Lateral loads applied along one (11)
or both axes to produce overturning moments due to their From geometry of tetrahedron (see Figs. 2 and 3) the following
eccentricity above the bottom of the footing is also included in equation for the remaining contact pressure qd can be obtained
the formula. When moments are transferred to the footing, they in terms of the other corner pressures.
may be produced by a resultant load P located eccentrically at
distances ex and e y from the centroid of the base of the footing q d = q b + q c − qa . (12)
shown in Fig. 3. The overburden may be introduced through the Substituting Eqs. (9)–(11) into Eq. (12) produces the remaining
footing weight that is also included in the presented formula. qd equation; that is,
The solutions presented here can be used directly to
estimate a footing contact area for all of the introduced load qd
combinations; no limit of kern area (L/6, B/6) needs to be −6L(Mx + h H y ) + B(−6h Hx − 6M y + L Pc ) + B 2 h L 2 γ
checked since it is already included in the formula that is based = .
B2 L2
on the bending theorem [15]. (13)
In Fig. 3, the total vertical load including the central column
load Pc and the weight of footing is shifted to positions ex and Eqs. (9)–(11) and (13) are referred as flexural formulae in
e y from centroidal column axes to provide the total bending foundation engineering for corner pressures under a rectangular
moments on both axes. The substitution of an eccentric load for footing and this is another way of driving the flexural formulae
the real moments simplifies the computations. rather than the bending theory [15].
According to the x and y axes for pressure block, the The solution of Eqs. (6)–(8) and (12) in terms B, L, qb and
equations for the eccentric load positions are as follows: qc yields the following equations:
qc
L L M y + h Hx
x̄ = − ex = − (4)
q
Pc2 (qa + qd ) − Pc2 (qa − qd ) − 72(h H y + Mx )(h Hx + M y )Pc (qa + qd − 2hγ )3
2 2 Pc + B Lhγ =
2Pc2
B B Mx + h Hy
ȳ = − e y = − . (5) (14)
2 2 Pc + B Lhγ
qb
Equating the total vertical load to the volume of pressure block Pc2 (qa + qd ) +
q
Pc2 (qa − qd ) − 72(h H y + Mx )(h Hx + M y )Pc (qa + qd − 2hγ )3
and those eccentric load distances to the centre of gravity =
2Pc2
distances of pressure block yields the following equations: (15)
B L(qb + qc ) L
Pc + B Lhγ = (6) q
2 Pc2 (qa − qd ) + Pc2 (qa − qd ) − 72(h H y + Mx )(h Hx + M y )Pc (qa + qd − 2hγ )3
=
L(4qb + 3qc − qa ) L M y + h Hx 6(h H y + Mx )(qa + qd − 2hγ )2
= − (7) (16)
6(qb + qc ) 2 Pc + B Lhγ
H.M. Algin / Engineering Structures 29 (2007) 1128–1134 1131

B
q
The above formula implies that these solutions can only be used
Pc2 (qd − qa ) + Pc2 (qa − qd ) − 72(h H y + Mx )(h Hx + M y )Pc (qa + qd − 2hγ )3 in practice at the present of biaxial bending about both axes.
= .
6(h Hx + M y )(qa + qd − 2hγ )2 Eqs. (21) and (22) occasionally have a functional
(17) discontinuity providing complex numbers. The complex
number is viewed as the following condition occurs in the
Now, the qd value that provides minimum footing area can be equations.
determined from analytical optimization of the whole footing
area (B × L). Eqs. (16) and (17) are multiplied, the first 972K < Pc3 . (26)
derivation of resulting equation is taken with respect to qd and
equated to zero. This is viewed as a point in the complex plane and it is regarded
as a point of discontinuity in the complex domain. These points
∂(B × L) ∂(Eq. (17) × Eq. (16)) in the domain are usually assumed to be absent, or they are
= = 0.
∂(qd ) ∂(qd ) explicitly excluded in order to have a continuous function
Since the second derivative produces positive argument the first on a disconnected real domain. In this case the insignificant
derivative can be used to minimise the footing area. The above imaginary part in the complex number can be removed and
equation can be solved in terms of the radical root of qd as the remaining real part of the solution accurately satisfies
follows; the conditions described above. However, in the case that the
complex number from the solution of Eqs. (21) or (22) needs to
Pc2 (qa − qd ) − 72(h Hy + Mx )(h Hx + M y ) be avoided, then the following equation can be used instead of
× Pc (qa + qd − 2hγ )3 = 0. (18) Eq. (23):
 p   
The solution of the equation above in terms qd produces KPc + 3 972K 4 + R1 Cos R32
the following required qd value which provides the minimum U= √ p (27)
footing area. 3K 6 972K 4 + R1

216K h + C
 where:
qd = hγ (19)  
216K h R1 = K 3 −972K + Pc3

where: √ 
3R1
R2 = ArcTan .
864KP3c − Pc6 54K 2
C = −216K + Pc3 + − Km (20)
Km Eq. (27) is derived from Eq. (23) which is rearranged by

eliminating the complex numbers in accordance with the
K m = −279 936K 2 Pc3 + 1296KP6c − Pc9 condition in Eq. (26) that generates the complex numbers.
q 1/3 It is rearranged to obtain a mathematical form in complex
+ 5184 3K Pc (972K − Pc )
3 6 3 coordinates, and then, Eq. (27) is obtained by taking the real
term of the complex expression and replacing the complex
K = (h Hy + Mx )(h Hx + M y )(qa − hγ ) arguments with the trigonometric angular expressions. If the
K h = (h Hy + Mx )(h Hx + M y )(hγ ). complex number viewed in Eq. (20), then Eq. (28) may be used
instead and it is derived in the same method described above.
Eq. (19) is substituted into Eqs. (16) and (17). Therefore, the    √ q  
simplified form of this final manipulation gives the following 1 5184 3 K 3 Pc6 972K − Pc3
C = Cos  π − Tan−1   
required B and L equations of minimum footing area subjected 3 279 936K 2 Pc3 − 1296KP6c + Pc9
to the biaxial bending about the both axes:
864KP3c − Pc6
 
L = U (h Hx + M y ) (21) × −216K + Pc3 + − Kn (28)
Kn
B = U (h Hy + Mx ) (22)
where:
where:    
r K n = Pc6 80 621 568K 3 Pc3 − 972K
√ 2
54K 2 + 3K 3 (972K − Pc3 )
3
3
p
3KPc +  2 1/6
U= r   (23) 2 3 6
+ 279 936K − 1296KPc + Pc .
9K 3 54K 2 + 3K 3 (972K − Pc3 )
3
p
Fig. 4 shows the graphical representation of Eqs. (21) and
K = (h Hy + Mx )(h Hx + M y )(qa − hγ ). (24) (22) in terms of various design values. Minimum footing area
values and dimensions can be obtained directly from Fig. 4
The ratio between the minimum footing dimensions can be
which can be used as a simple design chart for a rectangular
obtained by dividing Eq. (21) to Eq. (22); that is,
footing under two-way eccentricity.
B h Hy + Mx Eqs. (21) and (22) can be adapted to the all possible sign
= . (25)
L h Hx + M y convention variations of applied moments in the case the sign
1132 H.M. Algin / Engineering Structures 29 (2007) 1128–1134

Fig. 4. The design chart for minimum footing area under two-way eccentricity.
H.M. Algin / Engineering Structures 29 (2007) 1128–1134 1133

Table 1
The numerical comparison of the results from the classical methods, the genetic optimization and the results obtained from Eqs. (21), (22) and (19)

and qd (see Eqs. (9)–(11) and (13)) are calculated by ensuring


that all of the conditions [15] ( f a ≥ qa , qa ≥ qb , qa ≥ qc , qa ≥
qd , qa > 0, qb > 0, qc > 0, ex ≤ L/6 and e y ≤ B/6) are
satisfied (see Table 1). Table 1 also provides results from the
genetic algorithm carried out for these given examples and it
shows that the presented formulae provide the minimum results
even better than genetic optimization.
As demonstrated in Fig. 4 the same results can also be
obtained from the design chart. It initially conforms for the
following example that the given h and γ (related to the
footing depth) values are sufficient to produce the positive qd
value and then it continues to determine the required minimum
dimensions under the full compression. In terms of the user-
defined h and γ values it is sometimes impossible to have a
full compression; many examples confirmed that the formula
Fig. 5. The application of Eqs. (21) and (22) in terms of varying sign also provides valid results for those cases. However, if the full
conventions of the moments. compression is required in terms of user-defined h and γ values
of moments are dissimilar than that is adopted in the derivation then Fig. 4 provides this by checking qd / hγ value and thus qd
of the equations. Fig. 5 explains how Eqs. (21) and (22) can is positive.
be applied for those cases without any modification. It shows Problem No. 1. Calculate the minimum estimated footing area
that the position of maximum compression zone varies in terms by using the data [4] (Pc , Mx , M y , Hx , Hy , h, γ , f a ) given in
of the sign of moments and the sign of moments is always Table 1.
positive in terms of the position of maximum compression Solution. The required minimum dimensions (B and L) can be
zone. For simplicity, it may be pointed out that the maximum obtained directly from Eqs. (21) and (22) by introducing the
compression zone and corner names move about clockwise above values of Pc , Mx , M y , Hx , Hy , h, γ and qa (= f a ).
from Fig. 5(a) to (d). B = 1.99 m and, L = 3.98 m, so, Area = B × L = 7.91 m2
which provides 0.7 m3 less footing volume than the original
3. Numerical examples and discussion result.
Some of the following typical examples were originally Problem No. 2. Design the dimensions of the minimum footing
solved by classical methods and already published else- area for a hammerhead bridge pier by using the data [6]
where [4,6]. They are presented here to demonstrate the valid- (Pc , Mx , M y and f a ) given in Table 1.
ity of Eqs. (21) and (22) as well as illustrating how the pro- Solution. Eqs. (21) and (22) provide the following results.
posed formulae can be applied in practice. All of the original B = 3.73 m and L = 4.65 m.
results and those obtained using the presented formulae in this Above confirms that unlike the design configuration adopted
paper are summarized in Table 1 for comparison. In these so- in the previous works [7,8] the corner load intensities might
lutions, maximum pressure qa and other corner pressure qb , qc have different values whilst satisfying the all of the conditions
1134 H.M. Algin / Engineering Structures 29 (2007) 1128–1134

without setting one corner of the pressure block to zero contact including the nonlinear stress-strain behaviour of the soil and
pressure and without equalizing the remaining corner values the stiffness behaviour of the footing.
to the half of the maximum pressure (see Table 1). Thus, this
enables the presented formula to be applied to more general Acknowledgements
design cases without any geometrical presumptions.
The other examples are solved with the same method and The authors would like to thank Dr Paki Turgut for the GA
summarized in Table 1 which demonstrates that the pressure optimization results given in Table 1 and Zeynep Yilmaz for her
and limit of kern area values satisfy the conditions required. helpful comments on the preparation of Fig. 4.
This suggests that the better results can be obtained by using
the presented formulae in which there is no need for any References
initial footing area dimensions and performing time consuming
[1] Das BM. Principles of foundation engineering. 4th ed. New York: PWS
iteration process, and also there is no need for checking the Publishing; 1999.
corner pressure and the limit of kern area values to ensure that [2] Macginley TJ, Choo BS. Reinforced concrete-design and theory and
the conditions are satisfied. examples. 2nd ed. London, New York: Taylor & Francis; 2001.
[3] Mosley WH, Bungey JH, Hulse R. Reinforced concrete design. 5th ed.
4. Summary and conclusion London: Antony Rowe Ltd; 1999.
[4] Celep Z, Kumbasar N. Reinforced concrete structures. 2nd ed. Istanbul
The dimensions of a footing area subjected to the (Turkey): Evrim Publishing; 1996.
[5] Ersoy U. Reinforced concrete slabs and footings. 1st ed. Ankara (Turkey):
combination of biaxial bending and column load have been
Evrim Publishing; 1995.
formulated by disallowing the tension to be developed between [6] Jarquio R, Jarquio V. Design footing area with biaxial bending. ASCE
the footing and elastic medium interface. In most design cases Journal of Geotechnical Engineering 1983;109(GT10):1337–41.
the tension between the footing and elastic mass is not required [7] Jarquio R, Jarquio V. Vertical stress formulas for triangular loading.
to develop, so in this case the whole footing area is of full ASCE Journal of Geotechnical Engineering Division 1984;110(1):73–8.
compression. This arrangement is also employed in this paper. [8] Vitone DM, Valsangkar AJ. Stresses from loads over rectangular areas.
ASCE Journal of Geotechnical Engineering Division 1986;112(10):
However, the solution technique presented here may be used
961–4.
in future for any type of linear load intensities assumed to [9] Boussinesq J. Applications des Potentials a I’Etude de I’Etude de
be developed under a rectangular footing. The numerical and I’Euilibre et Mouvement des Solides Elastiques. Paris: Gauthier-Villard;
arithmetical solutions of a typical present day design cases for 1885.
the linearly distributed pressure regime have been compared [10] Gray H. Stresses and displacements from loads over rectangular area.
with the results from the general formula presented in this ASCE Civil Engineering 1943;13(5):227–9.
[11] Dempsey JP, Li H. A flexible rectangular footing on a Gibson soil:
paper. This paper concludes that the formula developed is valid.
required rigidity for full contact. International Journal of Solids and
Several tests have demonstrated that subsets of the general Structures 1995;32(3–4):357–73.
solutions provide solutions that are identical to the specific [12] Algin HM. Stresses from linearly distributed pressures over rectangular
solutions previously developed for those subsets. This suggests areas. International Journal of Numerical and Analytical Methods in
that the formula is valid for all of the more general cases of two- Geomechanics 2000;24:681–92.
way eccentricity. In this respect, it represents a significant step [13] Algin HM. Vertical stress formula for pressure over rectangular areas.
Géotechnique 2001;51(8):719–22.
forward in estimating the dimensions of a rectangular footing
[14] Newmark NM. Simplified computation of vertical pressures in elastic
area resting upon an elastic mass. The presented formula and foundations. University of Illinois Engineering 1935; Experiment Station,
the design chart can be used for regular rectangular footing Circular No. 24, Illinois.
design problems. However, for more exacting cases other [15] Popov EP. Engineering mechanics of solids. 2nd ed. New York:
methods are recommended, such as the finite element method Simon&Schuster; 1998.

You might also like