|
MacroFlow is ideally suited for system-level thermal design during the Conceptual Design stage. Its
object-oriented nature enables quick construction of flow networks of cooling systems and the powerful solution method enables rapid analysis. Thus, many different system layouts, what if
studies, and contingencies such as fan failure can be evaluated very quickly for arriving at few good system-level design early in the design cycle.
MacroFlow is a productivity tool. Its use results in significantly shorter design cycles, better product quality, and
reduces the time to market.
Construction of Flow Networks for Electronics Systems
Practical electronics cooling systems can be represented as a network of
components such as ducts, heat sinks, screens, filters, passages within card arrays, fans, bends, and tee junctions. Interconnections of these components correspond
to the paths followed by the coolant as it passes through the system. Flow and thermal characteristics of individual components are obtained from handbooks,
laboratory testing, and vender data. The emphasis of FNM is the analysis of the interaction among the components for determining the system performance.
Therefore, prediction of the details of flow and heat transfer within a component is not attempted.
Types of Cooling Systems Designed Using MacroFlow
MacroFlow is applicable to the design and analysis of system level cooling for
electronics used in: computers, data processing, telecommunications, military and commercial avionics, automotive and transportation equipment, consumer goods,
and medical applications. MacroFlow can be applied to open or closed systems, air or liquid cooling, and forced or natural convection for:
|
|
- Liquid cooled supercomputers
|
|
|
- Telecommunications cabinets
|
|
|
|
|
|
|
|
|
- Electronic navigation systems
|
|
|
- Space and satellite electronics
|
An Air-Cooled Server
A Liquid-Cooling System
Benefits and Limitations of MacroFlow
FNM offers a simple, quick, and accurate method for flow and thermal performance
of electronics systems. Some of the benefits it offers for system-level thermal design are described below.
- Evaluation of Competing Designs The strength of FNM is the analysis of
system-wide interaction of individual components. Thus, thermal performance of competing physical layouts of the system can be evaluated very quickly
and accurately through FNM analysis of corresponding flow networks involving different interconnections of the same set of system components.
- New Concepts for Design Improvements FNM analysis of a system provides a
clear overview of the flow and temperature distribution in the system. This is very useful not only in identifying the problem areas in the system but also in
generating ideas for design improvements (e.g. incorporation of flow balancing elements, addition of backup fans). Further, benefits of these improvements can also be quickly evaluated.
- "What If" Studies FNM analysis is ideally suited for determining the
magnitude of the impact on system performance under "what if" and contingency scenarios (e.g. fan failure and rise in ambient temperature).
- Complementary Use with CFD FNM and CFD complement and enhance each
other. Thus, CFD can be used for accurate determination of characteristics of nonstandard components (e.g. complex heat sinks) for use in FNM analysis.
Similarly, results of the FNM analysis of an entire system can be used to provide boundary conditions for a detailed analysis of part of a system (e.g.
card array) using CFD. FNM also enables focused use of CFD for the analysis of the most feasible system layouts.
The user should also be aware of the following limitations of this approach and use CFD where applicable:
- Component Temperatures FNM cannot predict the temperatures at the
component level. A detailed board-level thermal network or CFD analysis is necessary for this purpose.
- Flow Resistances The accuracy of FNM results depends upon the validity of the flow resistance correlations employed.
- Network Representation Flow network representations may not be accurate
or even possible for systems in which the flow paths are not well defined. An example of such a system would be an externally cooled sealed cabinet that
has large open spaces inside it. A network representation for the buoyancy driven flow inside such a system would be difficult to construct. Therefore,
FNM analysis cannot be used when the flow system cannot be represented as a network of identifiable flow paths.
The Enhanced Design Cycle
An enhanced design cycle that incorporates FNM in the early design stage, is shown
in the flow chart below. Use of FNM for Conceptual System Design significantly reduces the effort that is otherwise required for system-level thermal analysis. CFD
can then be used in a focused manner for detailed analysis of flow distribution and component temperatures in critical parts of a system or in specific competing
system designs. The proposed design cycle significantly shortens the time required for arriving at the final design and improves the quality of the product by enabling
the thermal engineer to explore more design options. Thus, use of FNM improves the productivity in the thermal design process and results in an optimum design cycle.
 |
Reference: Steinbrecher R., Radmehr A, Kelkar K.M., and Patankar S.V., "Use of Flow Network
Modeling (FNM) for the Design of Air-Cooled Servers", Proceedings, the Pacific Rim/ASME International
Intersociety Electronic and Photonic Packaging Conference (InterPack), vol.2, 1999, pp. 1999-2008.
|
