A first generation, one-dimensional predictive model is proposed for
designing heat sinks with fractal flow networks. A three-dimensional
computational fluid dynamics (CFD) model is analyzed as a means for validating
the model and identifying areas for improvement.
Two separate CFD models were developed. One was analyzed with
conjugate heat transfer whereas the other was not. For the conjugate heat transfer
model, heat flux was provided at a single surface, simulating a heat source. Energy
addition to the latter model, referred to as the non-conjugate model, was uniform to
all surfaces and was developed to assess the assumptions employed in the one-dimensional
model.
Both CFD models were run with and without variable properties and are
compared to results with a series of parallel channels with identical convective
surface areas. In all cases, with and without conjugate heat transfer and with and
without variable fluid properties, the fractal flow network showed lower maximum
surface temperatures than the straight channel network for identical pumping
powers. The pumping power, however, was determined assuming constant fluid
properties.
The variation in fluid viscosity with temperature was determined to have a
significant impact on the pressure distribution, which indicates that variable fluid
viscosity needs to be included in the one-dimensional model.
Also varied in the analyses were heat sink material, heat flux and flow rate.
Qualitative results show that temperature variations within the copper substrate are
less significant than in the stainless steel substrate. All analyses, including the one-dimensional
model, were restricted to laminar flow conditions. / Graduation date: 2002
| Identifer | oai:union.ndltd.org:ORGSU/oai:ir.library.oregonstate.edu:1957/29574 |
| Date | 29 November 2001 |
| Creators | Alharbi, Ali Y. |
| Contributors | Pence, Deborah V. |
| Source Sets | Oregon State University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
Page generated in 0.0015 seconds