Experimental study on the performance of a mechanical cooling tower fitted with different types of water distribution systems and drift eliminators

Water drift emitted from cooling towers is objectionable for several reasons, mainly due to human health hazards. Generation and control of drift depends mostly on the couple of elements water distribution system and drift eliminator. The configuration of these two components not only affects drift but also the cooling tower thermal performance. However, no references regarding the effect of the water distribution system on the cooling tower characteristic have been found in the reviewed bibliography. This paper presents an experimental investigation of the thermal performances of a forced draft counter-flow wet cooling tower fitted with a gravity type water distribution system (GWDS) for six drift eliminators and when no drift eliminator was fitted. The interaction between distribution system and drift eliminators is analyzed. Heat and mass transfer processes taken place in the cooling tower have found to be affected by the mass transfer coefficient and the exchange mass-heat area per unit of cooling tower volume. The comparison between the obtained results and those found in the literature indicates that the pressure water distribution systems type (PWDS) achieve better performances than the GWDS. Maximum averaged differences of 38.66% in terms of cooling tower performance have been obtained between the two water systems. The data registered in the experimental set-up were employed to obtain correlations of the tower characteristic. The outlet water temperature predicted by these correlations was compared with the experimentally registered values, obtaining a maximum averaged difference of ±1.61% for the water-to-air mass flow ratio correlation and ±0.95% for the water and air mass flow ratios.

► A variation in the water distribution system can increase cooling tower's thermal performance up to 40%.
► The presence of eliminators do not necessarily worsen the cooling tower's performance.
► The correlations predict the thermal performance of the cooling tower well (error < 0.95%).