The Colburn and fanning friction factors are computed using empirical correlation developed from experimental data covering a broad range of geometries and Reynolds numbers. For OSF, the Manglik & Bergels (1995) correlation is used.

The accuracy depends on the topology and geometry being investigated. Comparison with experimental data obtained by Kays & London shows that the Manglik & Bergles (1995) correlation for OSF fits over 95% of the dataset of geometries with maximum deviations of 20%.

The Area Goodness Factor (AGF), defined as the ratio j/f, is often used as an extended surface performance metric. The AGF is used to compare the required free flow area (A0) of given heat transfer surfaces, for given operating conditions, i.e., mass flow rate, temperature difference, and pressure drop. A geometry characterized by a higher AGF compared to another one indicates that, for a prescribed pressure drop and heat duty, that geometry will require a smaller frontal area and it will therefore yield a more compact design.

The validity range for the thermohydraulic performance calculation is reported on the x-axis of the chart, and is based on the experimental dataset used to develop the empirical correlation .Results outside this range should be used with caution.

The Colburn j factor (j = St * Pr^(2/3) = Nu * Re^(-1) * Pr^(-1/3)) is a dimensionless heat transfer coefficient. The Fanning friction factor (f) is a dimensionless measure of pressure drop: more specifically it approximates the slope of the pressure VS flow length in the heat exchanger core. Higher j values indicate better heat transfer, while lower f values indicate lower flow resistance.

The scaled mass goodness factor measures how much mass does the extended surface topology require to exchange a given amount of heat, per unit it of temperature difference between the corresponding fluid and the wall separating the two fluid streams. Therefore MGF = VGF/(rho_material×(1-phi)). Note that the scaled MGF* assumes a material density of 2700 and a fluid conductivity of 0.026 W/mK. Thus, MGF = MGF*×(k_fluid/0.026)/(2700/rho_material).

The VGF is a performance metric that quantifies the heat transfer rate achieved per unit volume V of the considered HX side, per unit of temperature difference between the corresponding fluid and the wall separating the two fluid streams.

You can calculate the heat transfer coefficient as h = j + Re * Pr^(-1/3) * k_f / d_h , where k_f is the fluid thermal conductivity (0.026 W/mK for air at 15<=T<=30 °C. The Prandtl number (Pr) is approximately 0.7 for air and other gasses, in most operating conditions.

Performance is calculated assuming air as the fluid (Pr=0.7, k_f=0.026 W/mK, density=1kg/m^3), and a series 3000 Aluminium alloy (k_e = 170 W/mK, density=2700 kg/m^3). Great question btw! With these info you can rescale the performance metrics for any other fluid or material.