Sara McAllister

Strategies for effective use of exergy-based modeling of data center thermal management systems

As power densities in data centers quickly increase, the inefficiencies of yesterday are becoming costly data center thermal management problems today. One proposed method to address the inefficiencies of state-of-the-art data centers is to use the concept of exergy. To this end, earlier investigations have used a finite-volume, uniform-flow computer model to analyze exergy destruction as a means of identifying inefficiencies. For this type of exergy-based program to be a useful engineering tool, it should: (i) be easy to set up, viz. establish grid size and impose system parameters; (ii) have a formulation that is solvable and numerically stable; (iii) be executable in reasonable time on a workstation machine with typical processor speed and memory; and (iv) model the physics with acceptable accuracy. This investigation explored specific strategies for achieving these features. This work demonstrates that optimally chosen computational strategies do enhance the usefulness of an exergy-based analysis program as an engineering tool for evaluating the thermal performance of a data center.

Effect of Environmental Variables on Flame Spread Rates in Microgravity

This paper reports 2D CFD-based computer modeling of opposed flow flame spread over thick samples of polymethylmethacrylate (PMMA). Model predictions are compared with experimental data from normal-gravity experiments at multiple forced flow velocities and KC-135 parabolic flight microgravity experiments. For the normal gravity experiments, good agreement between the model predictions and experimental data is obtained at one oxygen level, but flame spread rates at other oxygen levels are not well predicted. Of the four microgravity data points, the model underpredicts the spread rate of two of the data points by 35% or less. However, the model overpredicts the other two data points by almost a factor of two. Potential reasons for the discrepancies between the model predictions and the experimental data are discussed.

Thermodynamics of Combustion

Although combustion processes often involve chemical reactions that may be far from equilibrium, the equilibrium state provides a useful guide on the ultimate combustion state if sufficient time is given. Chemical compositions of the combustion products at equilibrium, heating value of a fuel, and flame temperature can be determined from thermodynamics. In comparison to the thermodynamics of a pure substance, the thermodynamics of combustion systems are complicated by the change of components during combustion. That is, the components in the final state are different from those in the initial state. With the introduction of enthalpy of formation, the general approach normally used to solve thermodynamic problems of a pure substance can be extended to combustion systems. The following topics will be discussed in this chapter: (1) properties of mixtures, (2) combustion stoichiometry, (3) heating values and enthalpy of formation, (4) adiabatic flame temperatures, and (5) equilibrium state (Cantera Program).

Ignition Delay of Combustible Materials in Normoxic Equivalent Environments

Material flammability is an important factor in determining the pressure and composition (fraction of oxygen and nitrogen) of the atmosphere in the habitable volume of exploration vehicles and habitats. The method chosen in this work to quantify the flammability of a material is by its ease of ignition. The ignition delay time was defined as the time it takes a combustible material to ignite after it has been exposed to an external heat flux. Previous work in the Forced Ignition and Spread Test (FIST) apparatus has shown that the ignition delay in the currently proposed space exploration atmosphere (approximately 58.6 kPa and 32% oxygen concentration) is reduced by 27% compared to the standard atmosphere used in the Space Shuttle and Space Station. In order to determine whether there is a safer environment in terms of material flammability, a series of piloted ignition delay tests using polymethylmethacrylate (PMMA) was conducted in the FIST apparatus to extend the work over a range of possible exploration atmospheres. The exploration atmospheres considered were the normoxic equivalents, i.e. reduced pressure conditions with a constant partial pressure of oxygen. The ignition delay time was seen to decrease as the pressure was reduced along the normoxic curve. The minimum ignition delay observed in the normoxic equivalent environments was nearly 30% lower than in standard atmospheric conditions. The ignition delay in the proposed exploration atmosphere is only slightly larger than this minimum. In terms of material flammability, normoxic environments with a higher pressure relative to the proposed pressure would be desired.

Droplet Evaporation and Combustion

Liquid fuels are widely used in various combustion systems for their ease of transport and storage. Due to their high energy content, liquid fuels are the most common fuels in transportation applications. Before combustion can take place, liquid fuel must be vaporized and mixed with the oxidizer. To achieve fast vaporization, liquid fuel is injected into the oxidizer (normally air) at high speeds. Soon after injection, the liquid fuel breaks up into droplets, forming a spray. Droplets then collide and coalesce, producing droplets of different sizes. Due to the high density of liquid fuel, the momentum of the liquid spray has a profound impact on local flow fields, creating turbulence and gas entrainment. In piston engines, the complexities of droplet combustion are further complicated by the occurrence of successive multiple transient events including gasification, ignition, flame propagation, and, ultimately, burn-out. As such, droplets can be considered the building block for providing fuel vapor in combustion systems. Understanding of single-droplet evaporation and combustion processes therefore provides important guidance in design of practical burners. Topics covered in this chapter include (1) droplet evaporation in both quiescent and convective environments, (2) droplet combustion, (3) initial droplet heating, and (4) characterization of droplet distributions.

Premixed Piston IC Engines

Internal combustion (IC) engines have been moving the industrial world for over three centuries. A type of IC engine is the spark ignition (SI) engine, where the fuel and oxidizer are premixed prior to entering the engine. Because of the high power density, low cost of production, and the vast infrastructure for gasoline, SI engines are ideal power platforms for passenger cars, small trucks, motorcycles, lawn mowers, and small electrical power generators. SI engines are robust and capable of producing high levels of power at wide speed ranges. Current opportunities for internal combustion engine development include efficiency improvement, novel fuel implementation, and pollution reduction. Topics in this chapter include: (1) principles of SI engines, (2) thermodynamic analysis, (3) discussion of the octane number, (4) fuel preparation, (5) ignition timing, (6) flame propagation, (7) modeling of SI engine combustion, (8) emissions and their control including catalytic converters, and (9) gasoline direct injection engines.

Non-premixed Flames (Diffusion Flames)

In many combustion processes, the fuel and oxidizer are separated before entering the reaction zone where they mix and burn. The combustion reactions in such cases are called “non-premixed flames,” or traditionally, “diffusion flames” because the transport of fuel and oxidizer into the reaction zone occurs primarily by diffusion. Many combustors operate in the non-premixed burning mode, often for safety reasons. Since the fuel and oxidizer are not premixed, the risk of sudden combustion (explosion) is eliminated. Chemical reactions between fuel and oxidizer occur only at the molecular level, so “mixing” between fuel and oxidizer must take place before combustion. In non-premixed combustion the fuel and oxidizer are transported independently to the reaction zone, by convection and diffusion, where mixing of the fuel and oxidizer occurs prior to their reaction. Often the chemical reactions are fast, hence the burning rate is limited by the transport and mixing process rather than by the chemical kinetics. Consequently, greater flame stability can be maintained. This stable characteristic makes diffusion flames attractive for many applications, notably aircraft gas-turbine engines. Topics covered in this chapter include: (1) a detailed description of a candle flame, (2) the structure of non-premixed laminar jet flames, (3) theoretical and empirical expressions for laminar jet flame height, (4) Burke-Schumann jet diffusion flames, (5) turbulent jet flames including liftoff height and blowout limit, and (6) a short discussion of condensed fuel fires.

Review of Transport Equations and Properties

The transport of heat and species generated by the chemical reactions is an essential aspect of most combustion processes. These transport processes can be described by the conservation equations commonly used in fluid and heat transfer analysis of engineering problems. Additional terms in the mass, momentum, and energy conservation equations account for the effects of the chemical reactions. This chapter briefly discusses heat and mass transfer, the equations governing combustion systems (conservation of mass, species, momentum, and energy), the normalization of the conservation equations, and simple expressions for the variation of viscosity, conductivity and diffusivity.

Effect of Pressure on Piloted Ignition Delay of PMMA

In order to reduce the risk of decompression sickness associated with spacewalks, NASA is considering designing the next generation of exploration vehicles and habitats with a different cabin environment than used previously. The proposed environment uses a total cabin pressure of 52.7 to 58.6 kPa with an oxygen concentration of 30 to 34% by volume and was chosen with material flammability in mind. Because materials may burn differently under these conditions and there is little information on how this new environment affects the flammability of the materials onboard, it is important to conduct material flammability experiments at the intended exploration atmosphere. One method to evaluate material flammability is by its ease of ignition. To this end, piloted ignition delay tests were conducted in the Forced Ignition and Spread Test (FIST) apparatus subject to this new environment. In these tests, polymethylmethacylate (PMMA) was exposed to a range of oxidizer flow velocities and externally applied heat fluxes. The ultimate goal is to determine the individual effect of pressure and the combined effect of pressure and oxygen concentration on the ignition delay. Tests were conducted for a baseline case of normal pressure and oxygen concentration, low pressure (58.6 kPa) with normal oxygen (21%). Future work will focus on low pressure with 32% oxygen concentration (space exploration atmosphere - SEA) conditions. It was found that reducing the pressure while keeping the oxygen concentration at 21% reduced the ignition time by 17% on average. It was also noted that the critical heat flux for ignition decreases in low-pressure conditions. Because tests conducted in standard atmospheric conditions will underpredict the flammability of materials intended for use on spacecraft, fire safety onboard at exploration atmospheres may be compromised.

Modeling Microgravity and Normal Gravity Opposed Flame Spread over Polymer/Glass Composites

Experimental data of opposed -flow flame spread obtained both in microgravity and normal gravity is used to validate the predictions of an analytical model for opposed -flow flame spread r ate. The model is based on a balance between heat -transfer controlled flame spread and kinetically controlled flame spread. The model predicts that the flame spread rate increases with opposed flow velocity, reaches a maximum, then decreases until extincti on. The fuels examined are thermally thick polymethylmethacrylate (PMMA) and polypropylene glass fiber composite (PPG). The microgravity flame spread experiments were conducted on NASA’s KC -135 research aircraft in the Forced Ignition and Flame Spread Test (FIST) at opposed -flow velocities lower than can be achieved in ground -based tests due to buoyancy -induced flow. In order to apply the analytical model to predict the experimental data, four model parameters need to be calibrated. Once this is done, excel lent agreement between the model and the data is achieved. The model predicts that for some conditions and fuels, flame spread in microgravity occurs in the heat transfer dominated regime subject to low velocity flow while that in normal gravity in the kin etically controlled regime. Consequently, there is a maximum in the flame spread rate at velocities lower than those generated by buoyancy in normal gravity. These results, which are supported by the experimental measurements, are relevant for fire safety in spacecraft since ground bas ed tests may under predict microgravity flame spread rates of materials intended for use on spacecraft.