his thesis presents the design and full-scale experimental and numerical investigation of mini-channel serpentine heat exchangers (MCHEs) for dispensing hot drinking water in commercial applications. The system dispenses hot water at a desired temperature and flow rate by controlling the electric power supplied to cartridge heaters and the water flow rate independently. All heat exchangers were fabricated from aluminium, which provides a faster transient thermal response compared to copper or stainless steel of equal volume. Numerical validation was performed using COMSOL Multiphysics, where conjugate heat transfer was solved with high mesh refinement. Transient efficiency was defined as the integral efficiency over a 40-second period. Among the first-generation designs, V01-2D3W (2 mm deep, 3 mm wide rectangular channel) showed the best transient efficiency, approximately 10% higher than V01-3D2W. A key design requirement was to deliver outlet water close to 99°C while keeping boiling minimal or suppressed. The Onset of Nucleate Boiling (ONB) position was evaluated for all tested geometries and configurations using non-uniform wall heat flux extracted from CFD simulations. The least boiling was observed at 500 W, 80°C, and 400 ml/min. In all other cases, boiling was significant and was suppressed either by increasing flow rate or by raising system pressure at constant flow rate. Increasing system pressure proved more effective, keeping wall temperatures below the ONB threshold predicted by the Sato correlation. Steady-state wall temperature at the last channel was used for numerical model validation, as the complex geometry limited access to other measurement points. Two new geometries were then introduced with a channel depth increased to 6 mm, while keeping the widths identical to the first-generation designs. The 6D3W design performed poorly, showing unstable boiling and early nucleation across all comparative tests. This was confirmed by apparent heat capacity phase-change CFD simulations, which showed wall superheats well above the ONB threshold in the last few channel bases. The 6D2W geometry was found to be a reasonable alternative to 2D3W and 3D2W when pressure drop reduction and boiling suppression are the primary concerns. Although heavier than first-generation MCHEs, the performance gap in transient tests can be largely closed by applying a pre-heating strategy, similar to that used in commercial water heaters. Observations from V01-3D2W, which operated with higher heat transfer coefficient and Nusselt number, motivated a second-generation redesign. The channel length was reduced to better align with the cartridge heater active length, eliminating cold spots at the channel bends. To maintain the same total channel length derived from the Sato correlation, the number of channels was increased from 10 to 12. Additionally, the number of cartridge heaters was increased from 6 to 7, so each element dissipates P/7 instead of P/6, resulting in more uniform heat flux distribution. Three new designs were evaluated: V02-3D2W-L, V02-3D2W-U, and V02-3D2.5W-U. The first two showed higher heat transfer coefficients and Nusselt numbers with lower thermal resistance and reduced boiling intensity. V02-3D2.5W behaved similarly to 6D3W, as the larger base and shorter wall caused wall superheat to exceed the ONB threshold. All designs were compared using the Performance Evaluation Criterion (PEC), which assesses whether the gain in heat transfer justifies the penalty in pressure drop. This metric gave consistent agreement with Nusselt number trends and confirmed that V02-3D2W-U and V02-3D2W-L outperformed all other designs, achieving a PEC factor of approximately 2. Design Recommendations: Minimum pressure drop / Early ONB / Lower flow stability / Large CHF margin → V01- 6D3W. This design produces lower wall heat flux and gentler boiling, but has the lowest thermal inertia and a lower heat transfer coefficient, resulting in a moderate transient response. Balanced pressure drop and high heat transfer density → V01-6D2W. This design shares the same channel cross-sectional area to flow volume ratio as the first-generation designs, but benefits from a larger hydraulic diameter, reducing pressure drop. It ranks better than V01- MCHE in thermal inertia and offers a good all-round thermal performance. It is likely the best general-purpose upgrade. Maximum UA density / Fastest transient response → V01-2D3W and V02-3D2W (U and L). These designs achieve the highest channel area to flow volume ratio. However, V01-2D3W carries a high risk of flow instability, while V02-3D2W carries a moderate instability risk alongside a higher pressure drop.
Design and Optimization of a Mini Channel Serpentine Heat Exchanger with Non-Uniform Heat Flux for Water Flash Heating
CHALLA, GANGA RAJU
2026
Abstract
his thesis presents the design and full-scale experimental and numerical investigation of mini-channel serpentine heat exchangers (MCHEs) for dispensing hot drinking water in commercial applications. The system dispenses hot water at a desired temperature and flow rate by controlling the electric power supplied to cartridge heaters and the water flow rate independently. All heat exchangers were fabricated from aluminium, which provides a faster transient thermal response compared to copper or stainless steel of equal volume. Numerical validation was performed using COMSOL Multiphysics, where conjugate heat transfer was solved with high mesh refinement. Transient efficiency was defined as the integral efficiency over a 40-second period. Among the first-generation designs, V01-2D3W (2 mm deep, 3 mm wide rectangular channel) showed the best transient efficiency, approximately 10% higher than V01-3D2W. A key design requirement was to deliver outlet water close to 99°C while keeping boiling minimal or suppressed. The Onset of Nucleate Boiling (ONB) position was evaluated for all tested geometries and configurations using non-uniform wall heat flux extracted from CFD simulations. The least boiling was observed at 500 W, 80°C, and 400 ml/min. In all other cases, boiling was significant and was suppressed either by increasing flow rate or by raising system pressure at constant flow rate. Increasing system pressure proved more effective, keeping wall temperatures below the ONB threshold predicted by the Sato correlation. Steady-state wall temperature at the last channel was used for numerical model validation, as the complex geometry limited access to other measurement points. Two new geometries were then introduced with a channel depth increased to 6 mm, while keeping the widths identical to the first-generation designs. The 6D3W design performed poorly, showing unstable boiling and early nucleation across all comparative tests. This was confirmed by apparent heat capacity phase-change CFD simulations, which showed wall superheats well above the ONB threshold in the last few channel bases. The 6D2W geometry was found to be a reasonable alternative to 2D3W and 3D2W when pressure drop reduction and boiling suppression are the primary concerns. Although heavier than first-generation MCHEs, the performance gap in transient tests can be largely closed by applying a pre-heating strategy, similar to that used in commercial water heaters. Observations from V01-3D2W, which operated with higher heat transfer coefficient and Nusselt number, motivated a second-generation redesign. The channel length was reduced to better align with the cartridge heater active length, eliminating cold spots at the channel bends. To maintain the same total channel length derived from the Sato correlation, the number of channels was increased from 10 to 12. Additionally, the number of cartridge heaters was increased from 6 to 7, so each element dissipates P/7 instead of P/6, resulting in more uniform heat flux distribution. Three new designs were evaluated: V02-3D2W-L, V02-3D2W-U, and V02-3D2.5W-U. The first two showed higher heat transfer coefficients and Nusselt numbers with lower thermal resistance and reduced boiling intensity. V02-3D2.5W behaved similarly to 6D3W, as the larger base and shorter wall caused wall superheat to exceed the ONB threshold. All designs were compared using the Performance Evaluation Criterion (PEC), which assesses whether the gain in heat transfer justifies the penalty in pressure drop. This metric gave consistent agreement with Nusselt number trends and confirmed that V02-3D2W-U and V02-3D2W-L outperformed all other designs, achieving a PEC factor of approximately 2. Design Recommendations: Minimum pressure drop / Early ONB / Lower flow stability / Large CHF margin → V01- 6D3W. This design produces lower wall heat flux and gentler boiling, but has the lowest thermal inertia and a lower heat transfer coefficient, resulting in a moderate transient response. Balanced pressure drop and high heat transfer density → V01-6D2W. This design shares the same channel cross-sectional area to flow volume ratio as the first-generation designs, but benefits from a larger hydraulic diameter, reducing pressure drop. It ranks better than V01- MCHE in thermal inertia and offers a good all-round thermal performance. It is likely the best general-purpose upgrade. Maximum UA density / Fastest transient response → V01-2D3W and V02-3D2W (U and L). These designs achieve the highest channel area to flow volume ratio. However, V01-2D3W carries a high risk of flow instability, while V02-3D2W carries a moderate instability risk alongside a higher pressure drop.| File | Dimensione | Formato | |
|---|---|---|---|
|
GangaRajuChalla_PhD_Thesis_Final_26.03.26_PDF_A.pdf
accesso aperto
Licenza:
Creative Commons
Dimensione
14.56 MB
Formato
Adobe PDF
|
14.56 MB | Adobe PDF | Visualizza/Apri |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/385026
URN:NBN:IT:UNIPI-385026