A theoretical background to the Hertz contact theory can be found here.

# Line Contact (Cylindrical contact)

In case of two cylinders in contact (with radii ), as shown in Fig. 1, the Hertzian radius of contact under applied normal load is given by the following equation:

(1)

where is the length of the cylinders and is the reduced elastic modulus. Equivalent radius is given by the following relation:

(2)

The mean and maximum pressures are given by:

(3)

Corresponding Matlab code for Hertz solution can be found here. The online Hertz contact calculator can be found here.

# Point Contact (Spherical contact)

For the case of two spheres in contact as shown in Fig.3, the Hertzian contact radius is given by the following equation:

(4)

, with is given by the following relation:

(5)

The elastic approach (also know as rigid body approach) is given by the following expression:

(6)

The mean and maximum pressures are given by:

(7)

Corresponding Matlab code for Hertz solution can be found here. The online Hertz contact calculator can be found here.

# Elliptical Point Contact

For the case of two spheres in contact as shown in Fig.4. In this case the Hertzian contact is an ellipse and is described by major () and minor axes of the contact ellipse:

(8)

where .

(9)

The functions in the previous equations are approximated as follows:

(10)

The mean and maximum pressures are given by:

(11)

Corresponding Matlab code for Hertz solution can be found here. The online Hertz contact calculator can be found here.

I believe the formula for the b (width of contact) is incorrect in two ways, firstly if b is the whole width it contradicts the diagram of the two cylinders in contact which shows 2b. So if it is the full width it should be b=2*(sqrt((4*F*R)/(pi*L*E’))) with a 4 instead of a 2 in the formula, OR if b is the half width as implied by the diagram it should be b=(sqrt((4*F*R)/(pi*L*E’))

Hi Owen,

Thank you for a comment. So b is the half-width, as shown in Figure 2. The equation for b that you posted in your comment assumes a different definition of the reduced elastic modulus (1/E’=…, while the equations posted in the wiki, assume 2/E’=…, see its definition after equation 1). If you substitute this formula to the equation in the article, you will get your equation.