Simulating the performance of the 24 ft. installed MinWaterCSP axial flow fan

Air-cooled condensers (ACCs) are used in thermoelectric power plants as dry-cooling solutions which reduce a plant’s reliance on water to cool the working fluid. These condenser systems use arrays of large diameter axial flow fans to discharge heat to the ambient air. However, assessing the installed performance of these large axial flow fans remains a challenge, especially when correlating full-scale installed performance with small-scale standardized tests, due to geometric and scaling effects. Furthermore, in large ACC installations it is often not feasible to install the instrumentation required to analyse the fans installed performance and resulting flow field, which brings about more uncertainty regarding their installed performance. Computational fluid dynamics (CFD) modelling provides a means to address this gap, though conventional reduced-order fan representations often compromise accuracy. Explicit fan modelling, while computationally demanding, offers a more rigorous approach to capturing the complex flow phenomena that govern installed performance. This study develops and validates a periodic three-dimensional CFD model (P3DM) of the ISO Type A fan test facility. The model demonstrated strong correlation with experimental measurements of shaft power and static pressure rise, achieving RMSE values below 5% and coefficients of determination (R2) between 0.98 and 0.99, thereby confirming its reliability at small scale. Scaling analysis using fan affinity laws revealed deviations in static pressure rise predictions of 7.6–14.8% depending on tip gap size, while shaft power scaling remained robust with deviations typically below 0.4%. These findings highlight the sensitivity of pressure predictions to geometric scaling, in contrast to the relative stability of shaft power scaling. At full scale, the performance of a 24 ft axial flow fan was simulated in both the ISO Type A facility and the MinWaterCSP test facility. At the maximum flow rate achievable in the MinWaterCSP facility (317 m³/s), shaft power predictions showed close agreement between models, whereas static pressure rise was underpredicted by 11.7% relative to ISO Type A results. The study highlights constraints associated with pressure measurements between an ISO Type A test facility and the MinWaterCSP test facility which reflects a Type B configuration. Flow field analysis indicated that structural differences between the facilities significantly altered inlet velocity distributions, thereby impacting installed fan performance. Overall, the study demonstrates that explicit CFD fan modelling provides reliable predictions of shaft power across scales, but static pressure rise remains sensitive to scaling effects and facility configuration. The results underscore the limitations of standardized testing in predicting installed fan performance and emphasize the importance of facility-specific analysis when evaluating large-scale ACC fans.

Grantor dc:publisher

Stellenbosch : Stellenbosch University

Year dc:date.issued

2026

Author dc:creator

• Duckitt, Matthew Richard

Advisors dc:contributor.advisor

• Van der Spuy, S. J.
• Venter, A. J.

Language dc:language.iso

en

Repository record dc:identifier.uri

https://scholar.sun.ac.za/handle/10019.1/135792

OAI identifier oai:identifier

oai:scholar.sun.ac.za:10019.1/135792