Separation efficiency of hydrocyclones: theoretical previsions vs. experimental results