Many empirical models have been used widely for designing hydrocyclones in industry. These empirical models consist of a set of design equations. Most of the design equations are based on the correlations obtained experimentally, using dimensionless similarity numbers. These equations have their limitations due to the specific systems used for model development. Therefore, in many cases, they cannot be used with confidence to predict the design of an individual separator, or the overall plant configuration that is required to meet different user requirements. A modified version of the existing design equations presented by previous researchers was developed. The design equations, describing four fundamental parameters: pressure drop, flow recovery to underflow, corrected cut size and classification function, were modified by fitting with set of experimental data to obtain the system-specific constant parameters. These modified design equations were then called, the fine-tuned design equations. Four sets of fine-tuned design equations describing all the fundamental parameters were constructed and used to predict the separation performance of many hydrocyclone operational systems at the Dow Chemical Company, Texas, USA. The performance of these equations is evaluated, and the limitations of their use are discussed, thus providing useful insights into hydrocyclone design. (c) 2005 Elsevier B.V. All rights reserved.