Experimental investigation of local air-side heat transfer coefficients of heat exchanger fins via an absorption-based mass transfer method

It is challenging to measure local air-side heat transfer coefficients (HTCs) with high accuracy due to the complexity of the involved surface geometries. An absorption-based mass transfer method has been developed to obtain local air-side HTCs. The method relies on measuring convective mass transfer and applying the analogy between heat and mass transfer. The new method is based on the interaction of a coating material and a tracer gas. Upon absorption of the tracer gas, the acidic coating layer reacts with a color change. Based on the rate of the observed color change, the local mass transfer can be quantified on the surface of interest. The thin and uniform coating layer can be applied to polymer and metal surfaces with high geometric complexity. The accuracy and robustness of the experimental method have been validated on fundamental geometries, such as external flow across a flat plate, wedge flow, two-dimensional stagnation flow, and external flow across a cylinder. The differences of span-averaged HTCs from measurements are within 10% compared to the analytical solutions and measurements from the open literature at similar conditions except for the edges. Moreover, the robustness of the experimental method has been investigated under a wide range of conditions. According to the size of the images and samples, the resolution of local HTC measurements is about 6-9 µm which is sufficient for investigating surfaces with complex geometries. Local air-side HTCs of real-scale heat exchanger (HX) fins have been experimentally investigated by employing the newly developed mass transfer method. Two-dimensional air-side HTC distributions for different fin-and-tube HX geometries have been visualized and quantified for a wide range of face airflow velocities (0.5-5 m/s). The samples include different fin types (plain, wavy, louver, and slit), tube bundle arrangements (inline and staggered), tube diameters (9.52 mm and 12.7 mm), and the number of tube rows (3-8). The averaged HTCs have been calculated by taking the arithmetic mean of the local HTCs of each pixel on the surface. Therefore, the averaged HTCs and pressure drops from the current measurements have been used to verify the accuracy of the experimental results by comparing them with data from the open literature. Moreover, the degradation of the HTCs with increasing row number has been investigated experimentally for the first time using actual fin stock material. The results show HTC degradation is related to the tube arrangement, airflow velocity, and row number. The results provide insights to propose new correlations for low and high flow velocities taking HTC degradation effects into account. Finally, the local air-side HTCs from the experimental study have been employed to verify local air-side HTCs obtained from numerical simulations. By reviewing the findings of previous studies, disagreement between averaged HTCs obtained with numerical and experimental methods is not uncommon. However, without local HTC measurements, it is nearly impossible to explain the differences and improve the numerical methods. Therefore, the current study has conducted numerical CFD simulations of the tested fin-and-tube HXs by using Fluent. Some important conclusions have been obtained by comparing the local air-side HTCs that were obtained experimentally and numerically: wrong predictions of local HTCs are observed even though the averaged HTCs from the numerical method agree with the experimental results. The models fail to capture horseshoe vortices, underestimate the HTCs in the wake region of the tubes, and over predict the row-by-row HTC degradation. Therefore, these effects need to be carefully considered when relying on numerical methods to design new fin shapes. Moreover, the accuracy of the numerical model decreases when the complexity of the geometry increases. Nonetheless, the model accuracy becomes worse at higher airflow velocities when the turbulent model is employed. Oversized fin-and-tube HXs with multiple tube rows are often selected as a result of using the under-predicted averaged air-side HTCs from the numerical CFD simulations. Therefore, the author has proposed a corrective method to achieve better agreement between experimental results and the averaged HTCs from the numerical simulations for the plain fin-and-tube HXs with multiple rows.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Mechanical Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2022

Author dc:creator

• Che, Min

Contributors dc:contributor

• Elbel, Stefan
• Jacobi, Anthony
• Miljkovic, Nenad
• Elliott, Greg

• Convection mass transfer Heat and mass transfer analogy Coating Tracer gas Color change Convective heat transfer coefficient Experimental air-side HTC distribution Fin-and-tube heat exchanger multiple tube rows heat transfer correlation

Statement dc:rights

• Copyright 2021 Min Che

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/112987

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/112987