A new model reveals how centrifugal pump energy loss varies with flow rate, heat transfer and rotational speed.
What really drives energy inefficiencies in centrifugal pumps? It’s a question that challenges OEMs, engineers and system designers alike. While much attention is paid to pressure and torque, new research has shown that other forces (namely fluid temperature, rotational speed, and internal flow state) can play a significant and underappreciated role in system losses.
Dr Longxiang Hu, a researcher at the Department of Refrigeration and Cryogenic Engineering within the School of Energy and Power Engineering at Xi’an Jiaotong University, led the study, which models how centrifugal pump energy loss is affected by multiple operating conditions. Along with colleagues from the same department, Hu developed a robust thermodynamic entropy-based method to assess loss variation across different working scenarios. Their findings have implications for design optimisation, performance diagnostics, and energy recovery strategies across water, HVAC and process applications.
“The model builds a fuller picture of what’s happening inside the pump, especially when working conditions are variable,” Hu said. “It shows how the energy loss is not just a function of pressure or mechanical load.”
The team’s full results are published in Energy, with the complete study available here via DOI for engineers seeking the full analysis.
Losses rise with heat and flow variation
The study focused on a single-stage centrifugal pump system. By applying the entropy generation method across the pump’s components, including the impeller and volute, the researchers isolated contributions to energy loss from internal viscous friction, flow separation, and turbulence.
Under stable conditions, the system’s energy conversion efficiency remained relatively high. But when flow rates deviated from the design point, especially under lower flow or shutoff scenarios, centrifugal pump energy loss increased significantly. This was linked to internal recirculation and flow instability within the impeller.
Meanwhile, increases in fluid temperature were found to reduce viscosity, leading to higher shear and energy dissipation in boundary layers. The loss did not scale linearly, revealing that elevated temperature and rotational speed together had a compound effect on performance.
This matters for real-world installations, where pumps are rarely operated under textbook conditions. “As fluid properties and speeds vary, the pump’s internal flow structure changes,” Hu said. “These changes affect not just the flow rate but also the thermodynamic loss mechanisms.”
The model also revealed that entropy generation rates varied spatially. Regions near the tongue of the volute and the trailing edge of the impeller blades showed higher local losses under off-design conditions. For system designers, this could help guide the use of advanced materials, coatings or impeller modifications to reduce energy dissipation where it matters most.
Implications for diagnostics, control and system design
While many pump performance models treat energy loss as a static figure, this new method highlights how dynamic operating conditions can shift the loss landscape significantly. The findings offer engineers a more accurate way to predict and optimise centrifugal pump performance, especially for variable-load systems like HVAC circuits or process lines with intermittent flow.
“Understanding how losses behave under different speeds and temperatures helps inform better control strategies,” Hu said. “It also helps with early fault detection if loss patterns deviate from expected norms.”
This could open the door for smarter diagnostics that rely on real-time flow and temperature inputs to estimate system losses. Rather than relying solely on pressure differential or pump curves, asset managers could use thermal and flow data to detect inefficiencies and make timely adjustments.
There’s also a design benefit. OEMs working on new centrifugal pump geometries can use this modelling method to test prototypes under variable conditions without requiring extensive field trials. This is especially useful in applications where pumps must maintain high efficiency across a broad range of flow and pressure conditions.
The research offers a path toward more resilient and responsive pump systems—those that not only perform well at design point, but that adapt better when operating under load, heat or speed variation.
