Thermoacoustic systems (engines or refrigerators) convert any source of heat energy, including solar energy and waste heat, into electricity or cooling effect. These systems are reliable and durable as they operate with few or no moving parts and they employ environmentally-friendly gases without gaseous emissions. However, thermoacoustic systems suffer from many non-linearities that deteriorate the overall performance after a certain level such as streaming, turbulence generation, entrance effects and harmonic generation. This thesis focuses on first three of these non-linearities. Streaming is a second order steady flow that convects a certain amount of heat. The convected heat does not contribute to the thermoacoustic conversion process and hence it represents a loss. In this work, the effects of the natural convection flow on Rayleigh streaming have been investigated. The first objective of this work is to investigate the distribution of the axial mean velocity inside a simple standing-wave thermoacoustic engine using both Particle Image Velocimetry and Laser Doppler Velocimetry. The engine consists of a stack heated from one side whereas the other side is left uncontrolled. The velocity measurements cover the axial distance from the cold side of the stack to the termination of the resonator. Also, dynamic pressure and mean gas temperature measurements are conducted. Three different regions are observed and named the “cold streaming” region, the “hot-streaming region” and the “end-effects” region. In the cold streaming region, the measured mean velocity distribution agrees well with the theoretical expectation of Rayleigh streaming at low acoustic level. At high acoustic level, the measured quantities deviate from the theoretical expectations reported in the literature. Also, the size of the cold streaming region was found to decrease with the increase of the acoustic level. In the hot streaming region, where the measured wall temperature gradient is non-zero, the measured mean velocity distribution does not agree with the theoretical expectation for all acoustic levels. This discrepancy was found to be caused by the natural convection flow originated by the non-uniform temperature distribution of the resonator wall. In this work, the natural convection flow is decoupled from the acoustic streaming flow in order to measure the natural convection flow distribution inside the engine. The results reveal that there is a competition between acoustic streaming flow and the natural convection flow. This competition, at some acoustic levels, results in a zero mean axial velocity distribution. In the end-effects region, the mean flow velocity is disturbed by the vortex generation near to the stack. The size of each of these three regions is determined for different acoustic levels. As the flow inside the thermoacoustic systems has an oscillating nature, the study of the transition to turbulence in the oscillating flow is critical to understand the flow characteristics at high velocity amplitudes, which is the second objective of this work. In this work, the transition to turbulence in an oscillating flow has been studied at two different frequency ranges namely the sub-acoustic (low) frequency range (i.e. frequency ≪ 20 Hz) and the acoustic (high) frequency range (i.e. frequency ≥ 20 Hz). In the sub-acoustic frequency range, the transition to turbulence under the oscillating flow conditions inside a square duct is investigated experimentally. For this purpose, the oscillating flow is generated by a mechanical system known as the Scotch-Yoke mechanism that is able to provide an oscillating flow with wide range of amplitudes at low frequencies. The axial velocity profile is measured using Particle Image Velocimetry and two dimensionless parameters are used to describe the oscillating flow namely the Reynolds number and the Womersley number. At low Reynolds numbers, the measured axial velocity profile in the duct agrees reasonably well with the theoretical laminar velocity profile over the complete cycle; whereas at higher Reynolds numbers, the results show that the agreement is limited to the acceleration phase. The transition to turbulence process is identified by measuring the turbulence intensities. The turbulence intensities based on both velocity components at both the center of the duct and near to the viscous penetration depth increase as the Reynolds number is increased. Also, the cycle-average Reynolds stress is estimated. Beyond a certain Reynolds number, the cycle-average Reynolds stress experiences a sudden increase indicating transition to turbulence and hence the value of the critical Reynolds number can be determined. The estimated value of the critical Reynolds number, which equals to 500, complies with the literature. The work is then extended to the acoustic (high) frequency range. In the acoustic frequency range, the mechanical Skotch-Yoke system is replaced by two powerful loudspeakers operating at 180 deg out-of-phase to produce an oscillating flow with high velocity amplitudes at the resonance frequency of the system. The same methodology is used to investigate the transition to turbulence at high frequency range. The cycle-average Reynolds stress experiences a sudden increase near the wall at Reynolds number of 270. The vorticity fields are calculated from the measured 2-D velocity field. It is found that the vorticity value increases as the Reynolds number is increased. Also, the largest vorticity value is observed near the wall. As the Reynolds number increases the largest vorticity value shifts away from the wall during the deceleration phases in the acoustic cycle. The spatial energy density spectrum is calculated at different phases for different Reynolds numbers. Also, the cycle-average spatial energy density spectrum is calculated. The slope of the decay of the spatial energy at high wavenumbers was found to be nearly equals to the universal value of -5/3. As the thermoacoustic core of any thermoacoustic systems consists of a stack and heat exchangers, the study of the oscillating flow behavior in the vicinity of the stack is important, which constitutes the third objective of this work. The stack is usually modelled as a set of parallel plates. This work focuses on the effects of the plate-end shape on the oscillating flow morphology at high velocity amplitudes. Four different plate-end shapes namely rectangular, circular, 90O triangular and 30O triangular are placed inside an acoustic resonator. The temporal evolution of the vorticity field in the vicinity of the plates is investigated. The vortices originated at the beginning of the ejection stage (flow moves outwards the plates) and moves with the flow till they reach an axial distance nearly equals to one acoustic displacement amplitude. As the Reynolds number increases, the generated rounded-vortices around each plates transformed from two counter-rotating vortices attached to the plate into two elongated counter-rotating vortices. Also, the non-periodicity (cycle-to-cycle variations) of the oscillating flow in the vicinity of the plates is reported. As the Reynolds number increases, the non-periodicity of the flow does not change significantly. The 30O triangular plate-end shape reduces the non-periodicity of the oscillating flow near the plates at different axial locations whereas the other plate-end shapes have nearly similar values of non-periodicity.


Mechanical Engineering Department

Graduation Date


Submission Date

May 2018

First Advisor

Abdel-Rahman, Ehab

Committee Member 1

Serag Eldin, Mohamed Amr

Committee Member 2

Essawey, Abdelmaged Hafez Ibrahim


219 p.

Document Type

Doctoral Dissertation


The author retains all rights with regard to copyright. The author certifies that written permission from the owner(s) of third-party copyrighted matter included in the thesis, dissertation, paper, or record of study has been obtained. The author further certifies that IRB approval has been obtained for this thesis, or that IRB approval is not necessary for this thesis. Insofar as this thesis, dissertation, paper, or record of study is an educational record as defined in the Family Educational Rights and Privacy Act (FERPA) (20 USC 1232g), the author has granted consent to disclosure of it to anyone who requests a copy.

Institutional Review Board (IRB) Approval

Not necessary for this item


The pursuit of my PhD has been a life-changing journey and it would not have been completed without the support of many persons and entities. First and foremost, I would like to acknowledge my advisors for all the support and guidance they gave me. I am grateful to Dr. Ehab Abdelrahman who has been supportive to my career goals and who worked sincerely to secure the financial and academic support not only for me but also for the whole thermoacoustic research team. Also, I would like to say a very big “thank you” to Dr. Abdelmaged who worked actively to provide me a sincere academic, professional and personnel guidance. I am also grateful to him for the nice “walk and talk” time we spent together discussing both scientific and general-life issues. I would like to thank Dr. Ahmed Abdel-Rahman for the fruitful discussions we had at the beginning of my PhD research. I would like to thank my colleagues in the thermoacoustic research team for the time we spent together. Also, I am thankful to my friend Khaled Elbeltagy for the sleepless nights we were working together to catchup on the deadlines of the projects. I am grateful to the financial support received towards my PhD from Yousef Jameel PhD fellowship through the American University in Cairo. I am also grateful to the fund received from the Egyptian Academy of Scientific Research and Technology to undertake my PhD. Also, I would like to thank the European union for funding my work . I greatly appreciate the support received from the collaborative research conducted with Pprime Institute, Poitiers University. I am especially grateful to Dr. Helene Bailliet for the sincere academic guidance during my stay in France. This collaborative work would not have been possible without the financial support received from both the American University in Cairo and Pprime Institute. Last but not least, I would like to thank my family for all their love and support. For my parents who did sacrifice their time and money to raise me with a love of hard working which is the pillar for all my pursuits. Most importantly, I would like to thank my loving, encouraging and patient wife, Alaa, for the faithful support throughout all ups and downs of my PhD journey. Thank you.