Theses and Dissertations at Montana State University (MSU)
Permanent URI for this collectionhttps://scholarworks.montana.edu/handle/1/733
Browse
3 results
Search Results
Item Controlling the area expansion of a backwards centrifugal fan blade passage using the principles of a diffuser and computational fluid dynamics(Montana State University - Bozeman, College of Engineering, 2021) Michalson, Adam Jeffrey; Chairperson, Graduate Committee: Erick JohnsonCentrifugal Fans are widespread in today's modern built environment. While a few variations of these fans exist, backward centrifugal fans are an efficient economical option capable of producing the pressure and airflow required for many modern building systems. Even though fans have become necessary piece of building engineering to facilitate occupant health and comfort, fan design almost exclusively relies on approximations to equations that have not changed since the 1950s and can consume, on average, 15% of a building's electrical consumption. Additionally, the approximations made support the ease and low cost of manufacturability. The traditional centrifugal fan design is made from stamped metal parts creating a fan blade sandwich with the blades held between an inlet shroud and a backplate. This rectangular blade passage is where the fluid flows through and picks up tangential acceleration. However, since the 1950s, nearly all advancements in fan design have been through incremental changes that are made by individual companies, and these resulting designs and performance data remain proprietary. This research revisits the foundations of centrifugal fan design with more modern tools and utilizes the concept of the diffuser to strictly control the expansion of the blade passage to improve centrifugal fan efficiency. Computational fluid dynamics was used to evaluate the performance of the new design against a traditionally manufactured fan. Combining the diffuser concept with an elliptical profile for the blade passage better controls the uniformity of the velocity field and pressure gradients through the passageway, while also reducing turbulence. Simulations of the new design against the traditional approach to fan design show an increase of nearly 10% in total efficiency.Item Diffusion and diffusive exchange are sensitive to the structure of cartilage as measured by nuclear magnetic resonance(Montana State University - Bozeman, College of Engineering, 2017) Mailhiot, Sarah Elizabeth; Chairperson, Graduate Committee: Ronald K. June II; Nathan H. Williamson, Jennifer R. Brown, Joseph D. Seymour, Sarah L. Codd and Ronald K. June were co-authors of the article, 'T1-T2 correlation and biopolymer diffusion within human osteoarthritic cartilage measured with nuclear magnetic resonance' in the journal 'Applied magnetic resonance' which is contained within this thesis.; Sarah L. Codd, Jennifer R. Brown, Joseph D. Seymour and Ronald K. June were co-authors of the article, 'Pulsed gradient stimulated echo (PGSTE) NMR shows spatial dependence of fluid diffusion in human stage IV OA cartilage' submitted to the journal 'Magnetic resonance in medicine' which is contained within this thesis.; Fangrong Zong, James E. Maneval, Ronald K. June, Petrik Galvosas and Joseph D. Seymour were co-authors of the article, 'Quantifying NMR relaxation correlation and exchange in articular cartilage with time domain analysis' submitted to the journal 'Journal of magnetic resonance' which is contained within this thesis.; James E. Maneval, Ronald K. June and Joseph D. Seymour were co-authors of the article, 'Relaxation exchange in human OA cartilage impacts the observable T 2 relaxation rates' submitted to the journal 'Magnetic resonance in medicine' which is contained within this thesis.Osteoarthritis (OA) is the deterioration of the tissue on the surface of the articulating joints in mammals. OA is the progression loss of articular cartilage. OA affects 50% of people over age 65 and is the leading cause of workplace disability. There is no cure for OA and the state of the art treatment is joint replacement. One limitation for treating OA is the difficulty of diagnosing OA before tissue failure. Magnetic Resonance Imaging (MRI) is capable of detecting early pathologic changes to cartilage but challenges remain. The goal of this work is to evaluate how parameters, specifically relaxation and diffusion, used for creating imaging contrast in MRI are affected by disease in naturally occurring human osteoarthritis. Nuclear Magnetic Resonance (NMR) is utilized to measure the diffusion and magnetic relaxation in human OA cartilage samples. Diffusion Weighted Imaging (DWI) is a proposed imaging mechanism for diagnosing OA. The hypothesis is that fluid diffusion is faster in diseased tissue than in healthy tissue. We show that diffusion of fluid increases when cartilage is damaged by enzymes, such as during OA. We also show that the diffusion of fluid is donor specific in human OA cartilage. Diffusion of proteins in cartilage is also sensitive to enzyme degradation and donor as well as to the size and structure of the proteins in cartilage. These are complementary measures of the fluid and solid phase of cartilage. Relaxation weighted imaging is the most common way to image cartilage and is capable of measuring small structure changes due to OA. One limitation of this method is that reported relaxation rates vary between studies. We show that exchange, or motion of fluid, between the two sites of relaxation in cartilage alters the observed relaxation. Further, we show that the exchange rate is sensitive to donor and enzyme degradation. The results suggest that exchange rate is a sensitive measure of structure in cartilage and that relaxation should be cautiously interpreted when exchange occurs. Overall, this work shows that NMR and MRI are sensitive to the structure of cartilage and capable of detecting pathological damage to cartilage.Item A numerical study of diffusion of nanoparticles in a viscous medium during solidification(Montana State University - Bozeman, College of Engineering, 2016) Rahman, Kazi Mizanur; Chairperson, Graduate Committee: M. Ruhul AminIn the field of additive manufacturing process, laser cladding is widely considered due to its cost effectiveness, small localized heat generation and full fusion to metals. Introducing nanoparticles with cladding metals produces metal matrix nanocomposites which in turn improves the material characteristics of the clad layer. The strength of the laser cladded reinforced metal matrix composite are dependent on the location and concentration of the nanoparticles infused in metals. Hence, investigating the nanoparticle diffusion characteristics during solidification of laser cladding process is of prime interest. The governing equations that controls the fluid flow are standard incompressible Navier-Stokes and heat diffusion equation whereas the Euler-Lagrange approach has been considered for particle tracking. The mathematical formulation for solidification is adopted based on enthalpy porosity method. During the solidification process of liquid titanium, particle flow and distribution has been observed until the entire geometry solidified. A two dimensional numerical analysis has been performed to identify and track the silicon carbide nanoparticle diffusion in titanium. A numerical model implemented in a commercial software based on control volume method has been developed that allows to simulate the fluid flow during solidification as well as tracking nanoparticles during this process. The influence of the free surface of the melt pool has a high importance on the fluid flow as well as the influence of pure natural convection. Thus both buoyancy and Marangoni convection have been considered in terms of fluid flow in the molten region. A detailed parametric study has been conducted by changing the Marangoni number, convection heat transfer coefficient, different initial distribution of particles, and thermal boundary condition of bottom wall to analyze the behavior of the nanoparticle movement. Variation in particle's initial distribution along with different Marangoni number and solidification time results in a high concentration of nanoparticles in some portion of the geometry and lack of particles in rest of the geometry. High concentration of nanoparticles decrease with a decrease in Marangoni number. Furthermore, an increase in the rate of solidification time limits the nanoparticle movement from its original position which results in different distribution patterns with respect to the solidification time.