Frey, Yannick Fabian (2024)
Bubble Nucleation in Pulsating Heat Pipes.
Technische Universität Darmstadt
doi: 10.26083/tuprints-00028599
Dissertation, Erstveröffentlichung, Verlagsversion
Kurzbeschreibung (Abstract)
In this work, experimental, and numerical investigations are presented that aim to allow further insights into the operational mechanisms of pulsating heat pipes. The focus is on the link between necessary wall superheat for bubble nucleation and thermal resistance. Therefore, three transparent experimental setups were manufactured for testing that were geometrically identical but had different evaporator roughness manufactured by milling, sand blasting, and glass bead blasting techniques. The overall thermal resistance between evaporator and condenser was measured with R1233zd(E) as a working fluid and used for the evaluation of thermal performance. The results showed that decreasing filling ratio, increasing heat flow, operating temperature, and evaporator roughness influenced the thermal resistance beneficially. The influence of surface roughness was most pronounced for low heat flow. From confocal measurements of the evaporator surface structure and a newly introduced evaluation algorithm, bubble radius (cavity size) distribution profiles could be estimated for each surface structure. This allowed to apply the pool boiling nucleation equation to calculate theoretical required wall superheats for nucleation for each data point of the measurements. Based on the results, a relative nucleation threshold (RNT) was introduced that represents the quotient of theoretical required wall superheat for nucleation to the temperature difference from evaporator to condenser available due to thermal conduction in an inactive PHP. It was found that the thermal resistance of the PHP correlates directly with RNT, and that the correlation is valid for multiple heat flow, surface roughness values, and operating temperatures. The coefficients of the correlation depended on the filling ratio. Using a high-speed camera, the flow inside the channels of the evaporator section was recorded. In combination with an optical bubble detection and motion tracking algorithm, average bubble nucleation rates (number of bubble generations per second) in the evaporator area for the analyzed time intervals were derived. The data at 70% filling ratio showed a connection between RNT and the bubble nucleation rate. The maximum bubble nucleation rate was located at an RNT value of 0.3. At RNT close to zero and high values of RNT, the bubble nucleation rate was approaching zero. For RNT>0.3, the behavior could be explained as higher RNT values lead to a lower probability of bubble nucleation and, thus, a reduced bubble nucleation rate. The decrease in nucleation rate from RNT 0.3 to RNT zero could not be quantitatively explained by the experiments. Flow observations suggested that low values of RNT cause a decrease of liquid mass in the evaporator. As bubbles can only be nucleated if liquid is present, the bubble nucleation rate is decreased here, even though RNT is also reduced. Subsequently, 1D-simulations of the PHP were conducted to gain further insights. To obtain correct physical understanding of PHP function from the simulations, they had to first be validated in the intended parameter range. Consequently, the experimental setups were reconstructed in the simulations. It could be shown that the simulation yielded accurate thermal resistances at 70% filling ratio and various condenser temperatures, heat flow, and surface roughness values. However, at lower filling ratios the simulation did not produce accurate results. Therefore, only simulations at 70% filling ratio were further analyzed. It could be shown that heat transfer coefficients between the fluid and the channel walls, the flow oscillation frequencies, and the liquid mass fractions inside the evaporator area correlated directly with RNT, strengthening the plausibility of the correlation found in the experiments. By presuming that bubbles can only be nucleated, if liquid is present within the evaporator, the liquid mass fractions found in the simulations allowed to quantitatively explain the bubble nucleation rates in the experiments. Compendiously, it could be shown that the required wall superheat for nucleation has a major influence on the PHP performance. To allow for the use of the correlation with different channel geometries, the correlation was empirically extended by further simulation data to provide a feasible PHP design tool. For this, simulations with various numbers of channels and evaporator lengths were conducted and a modified version of the correlation was introduced, accounting for the total channel length within the evaporator area.
Typ des Eintrags: | Dissertation | ||||
---|---|---|---|---|---|
Erschienen: | 2024 | ||||
Autor(en): | Frey, Yannick Fabian | ||||
Art des Eintrags: | Erstveröffentlichung | ||||
Titel: | Bubble Nucleation in Pulsating Heat Pipes | ||||
Sprache: | Englisch | ||||
Referenten: | Stephan, Prof. Dr. Peter ; Marengo, Prof. Dr. Marco | ||||
Publikationsjahr: | 10 Dezember 2024 | ||||
Ort: | Darmstadt | ||||
Kollation: | xiv, 124 Seiten | ||||
Datum der mündlichen Prüfung: | 10 Juli 2024 | ||||
DOI: | 10.26083/tuprints-00028599 | ||||
URL / URN: | https://tuprints.ulb.tu-darmstadt.de/28599 | ||||
Kurzbeschreibung (Abstract): | In this work, experimental, and numerical investigations are presented that aim to allow further insights into the operational mechanisms of pulsating heat pipes. The focus is on the link between necessary wall superheat for bubble nucleation and thermal resistance. Therefore, three transparent experimental setups were manufactured for testing that were geometrically identical but had different evaporator roughness manufactured by milling, sand blasting, and glass bead blasting techniques. The overall thermal resistance between evaporator and condenser was measured with R1233zd(E) as a working fluid and used for the evaluation of thermal performance. The results showed that decreasing filling ratio, increasing heat flow, operating temperature, and evaporator roughness influenced the thermal resistance beneficially. The influence of surface roughness was most pronounced for low heat flow. From confocal measurements of the evaporator surface structure and a newly introduced evaluation algorithm, bubble radius (cavity size) distribution profiles could be estimated for each surface structure. This allowed to apply the pool boiling nucleation equation to calculate theoretical required wall superheats for nucleation for each data point of the measurements. Based on the results, a relative nucleation threshold (RNT) was introduced that represents the quotient of theoretical required wall superheat for nucleation to the temperature difference from evaporator to condenser available due to thermal conduction in an inactive PHP. It was found that the thermal resistance of the PHP correlates directly with RNT, and that the correlation is valid for multiple heat flow, surface roughness values, and operating temperatures. The coefficients of the correlation depended on the filling ratio. Using a high-speed camera, the flow inside the channels of the evaporator section was recorded. In combination with an optical bubble detection and motion tracking algorithm, average bubble nucleation rates (number of bubble generations per second) in the evaporator area for the analyzed time intervals were derived. The data at 70% filling ratio showed a connection between RNT and the bubble nucleation rate. The maximum bubble nucleation rate was located at an RNT value of 0.3. At RNT close to zero and high values of RNT, the bubble nucleation rate was approaching zero. For RNT>0.3, the behavior could be explained as higher RNT values lead to a lower probability of bubble nucleation and, thus, a reduced bubble nucleation rate. The decrease in nucleation rate from RNT 0.3 to RNT zero could not be quantitatively explained by the experiments. Flow observations suggested that low values of RNT cause a decrease of liquid mass in the evaporator. As bubbles can only be nucleated if liquid is present, the bubble nucleation rate is decreased here, even though RNT is also reduced. Subsequently, 1D-simulations of the PHP were conducted to gain further insights. To obtain correct physical understanding of PHP function from the simulations, they had to first be validated in the intended parameter range. Consequently, the experimental setups were reconstructed in the simulations. It could be shown that the simulation yielded accurate thermal resistances at 70% filling ratio and various condenser temperatures, heat flow, and surface roughness values. However, at lower filling ratios the simulation did not produce accurate results. Therefore, only simulations at 70% filling ratio were further analyzed. It could be shown that heat transfer coefficients between the fluid and the channel walls, the flow oscillation frequencies, and the liquid mass fractions inside the evaporator area correlated directly with RNT, strengthening the plausibility of the correlation found in the experiments. By presuming that bubbles can only be nucleated, if liquid is present within the evaporator, the liquid mass fractions found in the simulations allowed to quantitatively explain the bubble nucleation rates in the experiments. Compendiously, it could be shown that the required wall superheat for nucleation has a major influence on the PHP performance. To allow for the use of the correlation with different channel geometries, the correlation was empirically extended by further simulation data to provide a feasible PHP design tool. For this, simulations with various numbers of channels and evaporator lengths were conducted and a modified version of the correlation was introduced, accounting for the total channel length within the evaporator area. |
||||
Alternatives oder übersetztes Abstract: |
|
||||
Status: | Verlagsversion | ||||
URN: | urn:nbn:de:tuda-tuprints-285990 | ||||
Sachgruppe der Dewey Dezimalklassifikatin (DDC): | 500 Naturwissenschaften und Mathematik > 500 Naturwissenschaften 600 Technik, Medizin, angewandte Wissenschaften > 600 Technik |
||||
Fachbereich(e)/-gebiet(e): | 16 Fachbereich Maschinenbau 16 Fachbereich Maschinenbau > Fachgebiet für Technische Thermodynamik (TTD) |
||||
Hinterlegungsdatum: | 10 Dez 2024 13:51 | ||||
Letzte Änderung: | 11 Dez 2024 07:21 | ||||
PPN: | |||||
Referenten: | Stephan, Prof. Dr. Peter ; Marengo, Prof. Dr. Marco | ||||
Datum der mündlichen Prüfung / Verteidigung / mdl. Prüfung: | 10 Juli 2024 | ||||
Export: | |||||
Suche nach Titel in: | TUfind oder in Google |
Frage zum Eintrag |
Optionen (nur für Redakteure)
Redaktionelle Details anzeigen |