Conceptual thermal design

Conceptual thermal design

Strijk, R.

Brezet, J.C. (promotor)

Industrial Design Engineering

2008-04-29

Present thermal design tools and methods insufficiently support the development of structural concepts engaged by typical practicing designers. Research described in this thesis identifies the main thermal design problems in practice. In addition, models and methods are developed that support an efficient conceptual thermal design process, resulting in higher quality products. Finally, the applicability, efficiency and quality of applying the models in design is evaluated by means of experiments and interviews. In the preliminary stage of this research, an overview is given of the practical problems encountering thermal design in practice. This resulted from an overview of the tools and methods described in literature and a series of semi-structured interviews with practicing designers. This preliminary research results in three main conclusions. First, many designers are unfamiliar with heat transfer and thermal design theories. Such designers lack the basic knowledge to make conceptual design choices and evaluate how important temperature will be during design. Second, evaluating structural concepts for heat flow and temperature development is not supported by a standard approach. Finally, temperature measurements for mock-ups and functional models are crucial in practical thermal design in order to find reliable heat transfer coefficients. Based on these results, three "conceptual" approaches have been developed that address one of the three problems. Each approach consists of a thermal design model and method for applying the model in a design situation. Model 1 is a chart, combined with a rule-of-thumb, that can be used as a risk assessment tool by a designer who has no knowledge of heat transfer or thermal design. Model 2 is a software tool that can be used to compute the heat path for structural concepts and is useful for rough temperature estimation in an electronic product. Model 3 is a theoretical formula that can be used to calculate the total transient behavior in a device and can be expanded by applying more detail. Because the model is theoretically derived, the designer can easily adjust the formula to fit a specific situation. Boundary conditions, thermal resistance and thermal capacitance are derived from measurements. The result is a semi-empirical model that is valid for a specific application and can be used to evaluate design changes. The three models and application methods have been evaluated by means of experiments using novice designers and interviews with practicing design engineers who all have experience in the design of electronic products with heat problems. The evaluation study shows that thermal design during the conceptual phase can be improved through several means. Designers can use the graph and rule of thumb embedded in model 1. By using this model, the most important structural concept decision regarding thermal design can be made: Should active or passive cooling be used? In addition, the designer obtains some idea of the importance of the role that thermal design will play in the development process. If the design is critical, the designer can choose to contact a thermal design specialist during the beginning stages to help define an optimal structural concept. This saves time and will most likely result in a design that is less vulnerable to after-defect solutions. A second step should including having the designer gain some background in heat transfer so that the concepts behind models 2 and 3 can be understood and applied. For this, it is necessary to expand these concepts into a various range of solutions that will evaluate the heat path for several structural concepts. The approach is comparable to that used in mechanical design, which has been successfully applied for many years. However, before entering this stage, a tremendous amount of experience data and modeling must be completed. The scientific relevance of this thesis does not just propose and develop models and methods that can improve the generation of structural concepts in early design phases. It also shows, for the first time, the application of an approach that combines scientific literature and interviews with designers from the field with thermal design research. Such an approach, in which the systematic use of available theory and knowledge of experts in the field, results in suggestions that can lessen the time needed to complete the product development cycle by several weeks and control the reliability and quality of resulting products.

thermal design

http://resolver.tudelft.nl/uuid:b3d6bf31-1036-48ac-87f7-330a2d52715b

978-90-9022969-0

Institutional Repository

doctoral thesis

(c) 2008 R. Strijk