A new generation of concentrating solar power (CSP) technologies is under development to provide dispatchable renewable power generation and reduce the levelized cost of electricity (LCOE) to 6 cents/kWh by leveraging heat transfer fluids (HTFs) capable of operation at higher temperatures and coupling with higher efficiency power conversion cycles. The U.S. Department of Energy (DOE) has funded three pathways for Generation 3 CSP (Gen3CSP) technology development to leverage solid, liquid, and gaseous HTFs to transfer heat to a supercritical carbon dioxide (sCO2) Brayton cycle. This paper presents the design and off-design capabilities of a 1 MWth sCO2 test system that can provide sCO2 coolant to the primary heat exchangers (PHX) coupling the high-Temperature HTFs to the sCO2 working fluid of the power cycle. This system will demonstrate design, performance, lifetime, and operability at a scale relevant to commercial CSP. A dense-phase high-pressure canned motor pump is used to supply up to 5.3 kg/s of sCO2 flow to the primary heat exchanger at pressures up to 250 bar and temperatures up to 715 °C with ambient air as the ultimate heat sink. Key component requirements for this system are presented in this paper.
This paper captures guidelines for the design and operation of sCO2 systems for research and development applications with specific emphasis on single-pressure pumped loops for thermal-hydraulic experiments and implications toward larger sCO2 Brayton power cycles. Direct experience with R&D systems at the kilowatt (kW), 50 kW, 200 kW, and 1 megawatt thermal scale has resulted in a recommended work flow to move a design from a thermodynamic flowsheet to a set of detailed build plans that account for industrial standards, engineering analysis, and operating experience. Analyses of operational considerations including CO2 storage, filling, pressurization, inventory management, and sensitivity to pump inlet conditions were conducted and validated during shakedown and operation of a 200 kilowatt-scale sCO2 system.
This report describes the design, development, and testing of a prototype 100 kWt particle-to-supercritical CO2 (sCO2) heat exchanger. An analytic hierarchy process was implemented to compare and evaluate alternative heat-exchanger designs (fluidized bed, shell-and-plate moving packed bed, and shell-and-tube moving packed bed) that could meet the high pressure (≥ 20 MPa) and high temperature (≥ 700 °C) operational requirements associated with sCO2 power cycles. Cost, heat-transfer coefficient, structural reliability, manufacturability, parasitics and heat losses, scalability, compatibility, erosion and corrosion, transient operation, and inspection ease were considered in the evaluation. A 100 kWt shell-and-plate design was selected for construction and integration with Sandia’s falling particle receiver system that heats the particles using concentrated sunlight. Sandia worked with industry to design and construct the moving packed-bed shell-and-plate heat exchanger. Tests were performed to evaluate its performance using both electrical heating and concentrated sunlight to heat the particles. Overall heat transfer coefficients at off-design conditions (reduced operating temperatures and only three stainless steel banks in the counter-crossflow heat exchanger) were measured to be approximately ~25 - 70 W/m2-K, significantly lower than simulated values of >100 W/m2-K. Tests using the falling particle receiver to heat the particles with concentrated sunlight yielded overall heat transfer coefficients of ~35 – 80 W/m2-K with four banks (including a nickel-alloy bank above the three stainless steel banks). The overall heat transfer coefficient was observed to decrease with increasing particle inlet temperatures, which contrasted the results of simulations that showed an increase in heat transfer coefficient with temperature due to increased effective particle-bed thermal conductivity from radiation. The likely cause of the discrepancy was particle-flow maldistributions and funnel flow within the heat exchanger caused by internal ledges and cross-bracing, which could have been exacerbated by increased particle-wall friction at higher temperatures. Additional heat loss at higher temperatures may also contribute to a lower overall heat-transfer coefficient. Design challenges including pressure drop, particle and sCO2 flow maldistribution, and reduced heat transfer coefficient are discussed with approaches for mitigation in future designs. Lessons learned regarding instrumentation, performance characterization, and operation of particle components and sCO2 flow loops are also discussed. Finally, a 200 MWt commercial-scale shell-and-plate heat-exchanger design based on the concepts investigated in this report is proposed.
The Generation 3 Concentrating Solar Power (Gen3CSP) supercritical carbon dioxide (sCO2) coolant loop, typically referred to here as the `sCO2loop,' is designed to continuously remove heat from a primary heat exchanger (PHX) subsystem through a flow of sCO2 as a substitute for a sCO2 Brayton power cycle as shown in Figure 1-1. This system is designed to function as a pumped coolant loop operating at a high baseline pressure with a high degree of flexibility, stability, and autonomy to simplify operation of a Gen3CSP Topic 1 team Phase 3 pilot plant. The complete system includes a dedicated inventory management module to fill the main flow loop with CO2 and recovery CO2 during heating and venting operations to minimize the delivery of CO2 to the site.
The generation 3 concentrating solar power, or Gen3CSP, campaign seeks to de-risk and deploy a CSP pilot plant through three parallel project tracks focused on solid, liquid, and gas-phase primary heat transfer fluids. Although the components between the sun and the primary heat exchanger from the thermal storage system differ with each track, the supercritical carbon dioxide (sCO2) coolant system required to cool the primary heat exchanger in place of a complete power conversion system has very similar requirements regardless of the primary heat transfer fluid. In order to avoid duplicative efforts, this project will design, assemble, perform acceptance testing, and deploy a single sCO2 coolant system design meeting the needs of any Gen3CSP topic 1 pathway pilot plant design.
Compact heat exchangers for supercritical CO2 (sCO2) service are often designed with external, semi-circular headers. Their design is governed by the ASME Boiler & Pressure Vessel Code (BPVC) whose equations were typically derived by following Castigliano’s Theorems. However, there are no known validation experiments to support their claims of pressure rating or burst pressure predictions nor is there much information about how and where failures occur. This work includes high pressure bursting of three semicircular header prototypes for the validation of three aspects: (1) burst pressure predictions from the BPVC, (2) strain predictions from Finite Element Analysis (FEA), and (3) deformation from FEA. The header prototypes were designed with geometry and weld specifications from the BPVC Section VIII Division 1, a design pressure typical of sCO2 service of 3,900 psi (26.9 MPa), and were built with 316 SS. Repeating the test in triplicate allows for greater confidence in the experimental results and enables data averaging. Burst pressure predictions are compared with experimental results for accuracy assessment. The prototypes are analyzed to understand their failure mechanism and locations. Experimental strain and deformation measurements were obtained optically with Digital Image Correlation (DIC). This technique allows strain to be measured in two dimensions and even allows for deformation measurements, all without contacting the prototype. Eight cameras are used for full coverage of both headers on the prototypes. The rich data from this technique are an excellent validation source for FEA strain and deformation predictions. Experimental data and simulation predictions are compared to assess simulation accuracy.
Compact heat exchangers for supercritical CO2 (sCO2) service are often designed with external, semi-circular headers. Their design is governed by the ASME Boiler & Pressure Vessel Code (BPVC) whose equations were typically derived by following Castigliano’s Theorems. However, there are no known validation experiments to support their claims of pressure rating or burst pressure predictions nor is there much information about how and where failures occur. This work includes high pressure bursting of three semicircular header prototypes for the validation of three aspects: (1) burst pressure predictions from the BPVC, (2) strain predictions from Finite Element Analysis (FEA), and (3) deformation from FEA. The header prototypes were designed with geometry and weld specifications from the BPVC Section VIII Division 1, a design pressure typical of sCO2 service of 3,900 psi (26.9 MPa), and were built with 316 SS. Repeating the test in triplicate allows for greater confidence in the experimental results and enables data averaging. Burst pressure predictions are compared with experimental results for accuracy assessment. The prototypes are analyzed to understand their failure mechanism and locations. Experimental strain and deformation measurements were obtained optically with Digital Image Correlation (DIC). This technique allows strain to be measured in two dimensions and even allows for deformation measurements, all without contacting the prototype. Eight cameras are used for full coverage of both headers on the prototypes. The rich data from this technique are an excellent validation source for FEA strain and deformation predictions. Experimental data and simulation predictions are compared to assess simulation accuracy.
Flow maldistribution in microchannel heat exchanger(MCHEs) can negatively impact heat exchanger effectiveness.Several rules of thumb exist about designing for uniform flow,but very little data are published to support these claims. In thiswork, complementary experiments and computational fluiddynamics (CFD) simulations of MCHEs enable a solidunderstanding of flow uniformity to a higher level of detail thanpreviously seen. Experiments provide a validation data source toassess CFD predictive capability. The traditional semi-circularheader geometry is tested. Experiments are carried out in a clearacrylic MCHE and water flow is measured optically with particleimage velocimetry. CFD boundary conditions are matched tothose in the experiment and the outputs, specifically velocity andturbulent kinetic energy profiles, are compared.
ASME 2018 12th International Conference on Energy Sustainability, ES 2018, collocated with the ASME 2018 Power Conference and the ASME 2018 Nuclear Forum
This paper presents an evaluation of alternative particle heat-exchanger designs, including moving packed-bed and fluidized-bed designs, for high-temperature heating of a solardriven supercritical CO2 (sCO2) Brayton power cycle. The design requirements for high pressure (> 20 MPa) and high temperature (> 700 °C) operation associated with sCO2 posed several challenges requiring high-strength materials for piping and/or diffusion bonding for plates. Designs from several vendors for a 100 kW-thermal particle-to-sCO2 heat exchanger were evaluated as part of this project. Cost, heat-transfer coefficient, structural reliability, manufacturability, parasitics and heat losses, scalability, compatibility, erosion and corrosion, transient operation, and inspection ease were considered in the evaluation. An analytical hierarchy process was used to weight and compare the criteria for the different design options. The fluidized-bed design fared the best on heat transfer coefficient, structural reliability, scalability and inspection ease, while the moving packed-bed designs fared the best on cost, parasitics and heat losses, manufacturability, compatibility, erosion and corrosion, and transient operation. A 100 kWt shell-and-plate design was ultimately selected for construction and integration with Sandia's falling particle receiver system.
Flow maldistribution in microchannel heat exchanger(MCHEs) can negatively impact heat exchanger effectiveness.Several rules of thumb exist about designing for uniform flow,but very little data are published to support these claims. In thiswork, complementary experiments and computational fluiddynamics (CFD) simulations of MCHEs enable a solidunderstanding of flow uniformity to a higher level of detail thanpreviously seen. Experiments provide a validation data source toassess CFD predictive capability. The traditional semi-circularheader geometry is tested. Experiments are carried out in a clearacrylic MCHE and water flow is measured optically with particleimage velocimetry. CFD boundary conditions are matched tothose in the experiment and the outputs, specifically velocity andturbulent kinetic energy profiles, are compared.