Publications

Dish concentrators can produce highly concentrated flux for the operation of an engine, a chemical process, or other energy converter. The high concentration allows a small aperture to control thermal losses, and permits high temperature processes at the focal point. A variety of optical errors can influence the flux pattern both at the aperture and at the absorber surface. Impacts of these errors can be lost energy (intercept losses), aperture compromise (increased size to accommodate flux), high peak fluxes (leading to part failure or life reduction), and improperly positioned flux also leading to component failure. Optical errors can include small scale facet errors ("waviness"), facet shape errors, alignment (facet pointing) errors, structural deflections, and tracking errors. The errors may be random in nature, or may be systematic. The various sources of errors are often combined in a "root-mean-squared" process to present a single number as an "error budget". However, this approach ignores the fact that various errors can influence the performance in different ways, and can mislead the designer, leading to component damage in a system or poor system performance. In this paper, we model a hypothetical radial gore dish system using Sandia's CIRCE2 optical code. We evaluate the peak flux and incident power through the aperture and onto various parts of the receiver cavity. We explore the impact of different error sources on the character of the flux pattern, and demonstrate the limitations of lumping all of the errors into a single error budget. © 2010 by ASME.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Charles Andraka from Sandia Labs was given the Hoyt Clarke Hottel Award for his lifelong work in the concentrating solar power industry.

Abstract not provided.

Concentrating Solar Power (CSP) dish systems use a parabolic dish to concentrate sunlight, providing heat for a thermodynamic cycle to generate shaft power and ultimately, electricity. Currently, leading contenders use a Stirling cycle engine with a heat absorber surface at about 800°C. The concentrated light passes through an aperture, which controls the thermal losses of the receiver system. Similar systems may use the concentrated light to heat a thermochemical process. The concentrator system, typically steel and glass, provides a source of fuel over the service life of the system, but this source of fuel manifests as a capital cost up front. Therefore, it is imperative that the cost of the reflector assembly is minimized. However, dish systems typically concentrate light to a peak of as much as 13,000 suns, with an average geometric concentration ratio of over 3000 suns. Several recent dish-Stirling systems have incorporated reflector facets with a normally-distributed surface slope error (local distributed waviness) of 0.8 mrad RMS (1-sigma error). As systems move toward commercialization, the cost of these highly accurate facets must be assessed. However, when considering lower-cost options, any decrease in the performance of the facets must be considered in the evaluation of such facets. In this paper, I investigate the impact of randomly-distributed slope errors on the performance, and therefore the value, of a typical dish-Stirling system. There are many potential sources of error in a concentrating system. When considering facet options, the surface waviness, characterized as a normally-distributed slope error, has the greatest impact on the aperture size and therefore the thermal losses. I develop an optical model and a thermal model for the performance of a baseline system. I then analyze the impact on system performance for a range of mirror quality, and evaluate the impact of such performance changes on the economic value of the system. This approach can be used to guide the evaluation of low-cost facets that differ in performance and cost. The methodology and results are applicable to other point- and line-focus thermal systems including dish-Brayton, dish-Thermochemical, tower systems, and troughs. Copyright © 2008 by ASME.

Abstract not provided.

Abstract not provided.

Abstract not provided.

A Dish Stirling parabolic concentrator typically consists of a number of mirror facets that must be aligned to focus the concentrated sunlight on the engine receiver. An alignment strategy must be developed to deliver the energy uniformly to the receiver while maximizing system performance. Several criteria must be met in order to maximize the performance and lifetime of the system. The peak flux should be minimized at the receiver to extend life. This is accomplished by locally optimizing the mirror aimpoints, minimizing overlap of facet images. The energy delivered to each cylinder of a multi-cylinder engine should be balanced to maximize the power production capability of the engine. This is accomplished through globally optimizing the mirror aimpoints. Depending on dish geometry, both of these constraints will be met by moving the aimpoints of certain facets away from a single point at the center of the aperture. However, this often results in a larger aperture or more flux spillage. The larger aperture results in greater thermal and reflective losses from the receiver cavity. This paper proposes and demonstrates a novel approach to optimizing the alignment strategy while obeying these constraints. The method uses an approach similar to molecular dynamics to globally and locally distribute the power on the receiver, while imposing movement constraints at the aperture to limit the focal plane spot size. The method can also impose additional geometric constraints at the receiver plane to accommodate un-cooled surfaces. The method is explored and demonstrated on the Stirling Energy Systems 25kW dish Stirling system at Sandia National Laboratories. The approach provides a receiver flux distribution and power balance equal to the strategy developed by McDonnell Douglas in the early 1980's, but with an aperture size equal to that of the single aimpoint strategy. This should result in about a 1kW increase in power generated at rated conditions, with no additional cost, due to reduced thermal losses from the receiver. The method can be extended to other point-focus concentrating solar technologies. On a tower, the heliostat aiming strategy could be dynamically updated to accommodate flux profile needs, sun position, or maintenance in the field. Copyright © 2007 by US Government.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

The integration and approaches utilized in the various stages of the Advanced Dish Development System (ADDS) project are presented and described. Insights gained from integration of the ADDS are also discussed. The ADDS project focuses on development of a product that meets the needs of the remote power market and helps to identify key technology development needs that resulted in a system that is closer to commercialization. A pursuance of solving problems, a lack of fear of breaking things, and hands-on involvement by design engineers are the key components leading to rapid improvement of the project.

Abstract not provided.

Abstract not provided.

The development of liquid metal heat-pipes for use in solar powered Stirling engines has led to an in-depth analysis of heat-pipe wick properties. To model the flow of liquid sodium through the wick its two-phase permeability measurement is of interest. The permeability will be measured by constructing a test cell made up of a wick sample sintered to a manifold. Measuring the volumetric flow rate through the wick will allow for a determination of the wick's permeability as a function of pressure. Currently, simple estimates of permeability as a function of vapor fraction of a porous media are being used as a model to calculate the two-phase permeability. The above mentioned experiment will be used to test the existing formulas validity. The plan is to make use of a known procedure for testing permeability and apply those techniques to a felt-metal wick. The results will be used to verify and/or modify the two-phase permeability estimates. With the increasing desire to replace directly illuminated engines with the much more efficient heat-pipe apparatus it is inherently clear that the usefulness of known wick properties will make wick permeability design a simpler process.

Stirling-cycle engines have been identified as a promising technology for the conversion of concentrated solar energy into usable electrical power. In previous experimented work, we have demonstrated that a heat pipe receiver can significantly improve system performance-over a directly-illuminated heater head. The design and operating conditions of a heat pipe receiver differ significantly from typical laboratory heat pipes. New wick structures have been developed to exploit the characteristics of the solar generation system. Typically, these wick structures allow vapor generation within the wick. Conventional heat pipe models do not handle this enhancement yet it can more than double the performance of the wick. In this study, I develop a steady-state model of a boiling-enhanced wick for a solar heat pipe receiver. The model is used for design-point calculations and is written in FORTRAN90. Some limited comparisons have been made with actual test data.

The thermal conduction of a portion of an enhanced surface heat exchanger for a gas fired heat pipe solar receiver was modeled using the boundary element and finite element methods (BEM and FEM) to determine the effect of weld fillet size on performance of a stud welded pin fin. A process that could be utilized by others for designing the surface mesh on an object of interest, performing a conversion from the mesh into the input format utilized by the BEM code, obtaining output on the surface of the object, and displaying visual results was developed. It was determined that the weld fillet on the pin fin significantly enhanced the heat performance, improving the operating margin of the heat exchanger. The performance of the BEM program on the pin fin was measured (as computational time) and used as a performance comparison with the FEM model. Given similar surface element densities, the BEM method took longer to get a solution than the FEM method. The FEM method creates a sparse matrix that scales in storage and computation as the number of nodes (N), whereas the BEM method scales as N{sup 2} in storage and N{sup 3} in computation.

Stirling-cycle engines have been identified as a promising technology for the conversion of concentrated solar energy into usable electrical power. A 25kW electric system takes advantage of existing Stirling-cycle engines and existing parabolic concentrator designs. In previous work, the concentrated sunlight impinged directly on the heater head tubes of the Stirling Thermal Motors (STM) 4-120 engine. A Sandia-designed felt-metal-wick heat pipe receiver was fitted to the STM 4-120 engine for on-sun testing on Sandia`s Test Bed Solar Concentrator. The heat pipe uses sodium metal as an intermediate two-phase heat transfer fluid. The receiver replaces the directly-illuminated heater head previously tested. The heat pipe receiver provides heat isothermally to the engine, and the heater head tube length is reduced, both resulting in improved engine performance. The receiver also has less thermal losses than the tube receiver. The heat pipe receiver design is based on Sandia`s second-generation felt-wick heat pipe receiver. This paper presents the interface design, and compares the heat pipe/engine test results to those of the directly-illuminated receiver/engine package.