15kw Free Piston

7/28/2019 15kw Free Piston

1/189

5c?TDOE/NASA/0056-79/1NASA CR- 1 59587MTI 79TR47

(MASft-Cfi-159587) DESIGM SIOCJ OF A 15 kS 880-22767FBEB-PISICM SIIfiLING EliGIliE-IIIiEAfiALTEfiHATCS FC CISPEBSEC SCLAE ELEClfilCPOHEB SYSIEOS Final Befott, Sep. 1978 - DnclasAug. 197S (aechanical Techoology, Inc.) G3/44 1S516

DESIGN STUDY OF A 15 kW FREE-PISTONSTIRLING ENGINE LINEAR ALTERNATOR FORDISPERSED SOLAR ELECTRIC POWER SYSTEMSGeorge R. DochatStirling Engine Systems DivisionMechanical Technology Incorporated

y'A'^ ' '.; /? /.- ,-

A --J /;V

August 1979 ';;V

Prepared forNATIONAL AERONAUTICS AND SPACE ADMINISTRATIONLewis Research CenterUnder Contract DEN 3-56

forU.S. DEPARTMENT OF ENERGYDivision of Central Solar Technology


2/189

DOE/NASA/0056-79/1NASA CR-1 59587MTI 79TR47

DESIGN STUDY OF A 15 kW FREE-PISTONSTIRLING ENGINE LINEAR ALTERNATOR FORDISPERSED SOLAR ELECTRIC POWER SYSTEMSGeorge R. DochatStirling Engine Systems DivisionMechanical Technology IncorporatedLatham. New York I2II0

August 1979

Prepared forNATIONAL AERONAUTICS AND SPACE ADMiM STRATIONLewis Research CenterC'eveland. Ohio 44135Under Contract DEN 3-56

forU.S. DEPARTMENT OF ENERGYDivision of Central Solar TechnologyWashington. D.C 20545Under Interagency Agreement EX-76-A-29-1060


3/189

t. RMOrt No.ilASA C& - U9S87

2. Govtrnmtfli Acetttion No. 3. RwiOMDl's Cllla9 No.

4. Title ind Subtitl*OSStOI Stmt OF A 15 kV FBSS PISTON STIRLIHC EMCIilE - UMEAftALTEBHAXOR TOR DtSPERSED SOUR ELECTRIC POHBR STST5MS

S. Rtoort Oat*AugMC 1979

6. PtrforaiMig Orfininnon Codt

7. Authorli)G.Soehat, B.S.Chon, S.Shace, T.Marusak

8. Ptrformmg Or9>niz*iion Reoon Mo79TR4;

10 Work Unil No.9. ^trfornung Orgmiaiion Ntm* and Addrot

Mtchmieat Tachnoloc/ Inc.968 Albcny-SIwkar BoAdUchu. Htw >rk I2U0

U. Contract or Gnni No.DEN 3-56

t2. Soontonng Agtnev Nnw and AddrtnO.S. 0parcaanc of EaergyOtvtsloa of ContraX Solar Technologytfuhingcon, DC 20545

13. Tvot ot Rtoon and Ptnoo Covrdrioal Ropoct 9/78 - 8/79

14. Sooraonng Aq(ncv4ado RoporC So.DOE/NASA 00S6 - 79/1

IS. SuOOUnwntary NoMtFinal Reporc. prepared vmder Inceragancy Agreameac EX-76^-29-1060.Project Kanagcr. G.Oochac. Sclrllng Engine Syscans Divtaioa, Mechanical Technology Inc.. Lachaai.iIT

16. AostraaThis sctidy ceaulcad Is a conceptual design of a free piston solar Stirling engine'llnear alternatoruhich can be designed and developed to meet the requireaients of a oear-tem solar test bed enginewith Binifflua risks. The conceptual design was calculated co have an overall system efficiency of38Z and provide 15kV electric output. The free-piston engine design incorporates features such asgas bearings, close clearance seals and gas springs. This design is henaecically scaled to providelong life, reliability and aainteitance free operation. An lnplementation assessaienc study perfome^as pare ot this study Indicates chat the Cree-ptston solar Stirling engine-linear alcematrr :an bemanufactured at a reasonable price cost (direct labor plus naterial) of $2,500 per engine s.n pro-duction quanticles sf 25,000 units per year. Opportunity for st^nifloant reduction of cose is alsoIdentified.

17.


4/189

TABLE OF CONTENTS

SectionLIST OF FIGURES v

1.0 INTRODUCTION 1-11.1 Background l-l1.2 Objectives 1-21.3 Scope 1-21.4 Requirements 1-31.5 Susnoary o Results 1-5

1.5.1 Engine Conceptual Design 1-52.0 CONFIGURATION DEFINITION STUDIES AND PARAMETRIC ANALYSIS . 2-1

2.1 Task Objectives 2-12.2 Technical Approach 2-1

. 2.3 Heat Input Concepts 2-22.3.1 An Integrated Heat Receiver, Thermal Storage

and Heat Transport System Concept 2-22.3.2 Heat Input Concept Selection 2-22.3.3 Reconuaended Heat Inpuc Concept 2-7

2.4 Engine Design Point Selection 2-142.5 Concepts Selected for Parametric Analysis 2-19

2.5.1 Preliminary Concepts 2-192.5.2 General Trade-offs 2-252.5.3 Layout/Description of Selected Concepts. . . . 2-30

2.6 Linear Alternators 2-322.6.1 Selection of Types of Alternators 2-342.6.2 Reasons for Selecting Flux Reversing Type

Permanent Magnet Alternator 2-362.6.3 Two Configurations of Permanent MagnetAlternator 2-36

2.6.4 Parameters of Design 2-3.0WCOIVOM__5rrEHTm'^T'W*iK


5/189

TABLE OF CONTENTS (cont'd)

Sectl


6/189

TABLE OF CONTENTS (cont'd)

Section Pa^e4.6 Component System Design 4-10

4.6.1 Engine 4-104.6.2 Permanent Magnet Linear Alternator 4-16

4.7 System Control and Stability 4-254.7.1 System Stability 4-254.7.2 Engine/Alternator Control 4-314.7.3 Power Modulation by Engine Pressure Control. . A^S?4.7.4 Overall System Control 4-394.7.5 Typical Operating Profile 4-41

4.8 Engine/System Interface 4-444.8.1 Engine/GE Heat Receiver 4-444.8.2 Direct Heat Receiver Concept 4-464.8.3 Engine/Load Interface 4-504.8.4 Auxiliary Heat Source 4-50

REFERENCES R-1

mCHMicM Vllrl-iWJotiA*'^wi^^^^9^9immtmmi


8/189

LIST OF FIGURES CONT'D

Fliture Page3-4 Prime Cost Estimates for the Conceptual Design and a

Developed Engine/Alternator System * 3-20S-Aa Prime Cost Estimates for the Conceputal Design and a

Developed Engine/Alternator System for Production Quantitiesof 25,000 Per Year 3-22

3-5 Learning Curve Rate as a Function of Assembly Labor 3-233-6 15 kW Stirling Engine 91 Percent Learning Curve 3-283-7 Material/Labor Dollars as a Function of Volume 3-313-7a Material/Labor Dollars as a Function of Volume 3-323-8 Power Output Versus Charge Pressure for the 15 kW

Conceptual Design 3-393-9 Power Output Versus Gas Spring Rate Ratio 3-403-10 Variation of Engine Power as a Function of Piston

Stroke for the 15 kW Solar Conceptual Design 3-413-11 Power Output and Efficiency Versus Heater Wall

Temperature for the 15 kU Conceptual Design 3-423-12 Engine Power and Heat Transfer Ratios as a Function

of Scaling Factor 3-463-13 Large-Power Free-Piston Stirling Engine Conceptual Design . . . 3-514-1 15 kW Preliminary Conceptual Design (Dimensions in cm) 4-24-2 15 kW Free-Pibcon Stirling Engine - Alternator

Conceptual Design (Dimensions in cm) 4-64-3 15 kW Conceptual Design Solar Stirling 4-184-4 Effect of Magnet Thickness on Efficiency at Constant

Magnet Volumes 4-194-5 Effect of Slot Opening on Power C rput 4-214-6 Nomenclature for Permanent Magnet Linear Alternator 4-244-7 Permanent Magnet Linear Alternator Design ..... 4-274-8 Lumped Parameter Stability and Control Model 4-294-9 Effect of Overall Alternator Resistance on Stability 4-324-10 Solar System Operational Modes , 4-344-11 Transient Engine Response Mode II to Mode I Conversion 4-354-12 Transient Engine Response Mode II to Mode I Conversion

(Cont'd) 4-364-13 Transient Engine Response Mode 1 to Mode II Conver


9/189

LIST OF FIGURES CONT'D

Figure Page4-15 Pressure Control Response Diagram 4-424-16 Basic Engine Control System 4-434-17 Typical Operating Profiles 4-454-18 Stirling Engine/Heat Receiver Interface 4-484-19 Initial Direct Insolation Heater Head Concept 4-504-20 Direct Insolation Heater Concept 4-53

-xi-ncMMOioa*


10/189

1.0 INTRODUCTIONThis final report contains the results of a six-month study contract awardedby the NASA-Lewis Research Center to develop a conceptual design of a nominal15 kW electric solar Stirling engine. Conceptual designs were evaluated anddeveloped for both a free-piston and kinematic-type Stirling engine for smalldispersed solar powered applications in identical parallel path programs.This volume contains only the results for the free-piston Stirling enginestudy. A separate volume contains similar information for the kinematic-type Stirling engine.

The study performed configuration definition studies, including a detailedparametric evaluation of the selected concepts, with a final ranking of allattractive configurations. Upon selection of the best configuration from anoverall system viewpoint, the study developed a conceptual design of a near-term solar engine capable of providing a nominal 15 kW electric output. Par-alleling the conceptual design, an implementation assessment of the selecteddesign was performed defining the producibllity (cost), durability and growthpotential.

It is concluded that a Stirling engine offers high conversion efficienciesover other hear, engines (>.40 percent) for small solar powered applications.This high efficiency will enable system cosfs to be reduced. The free-pistonStirling engine driving a linear alternator offers the potential for longlife, inherent reliability and maintenance-free characteristics of a hermeti-cally-sealed power unit.

1.1 BackgroundAs part of the Solar Thermal Power Program in the Division of Solar TechnologyU.S. Department of Energy, studies and experiments are being conducted forCentral Power Systems application and Dispersed Power Systems Applications.In support of these systems, the Advanced Techno! ngy Rrnr.ch cf Llie aoove divi-sion is conducting research on advanced systems and subsystems for increasedefficiency and decreased cost of electricity from solar energy input. One ofthese systems is referred to as a roint-focusing, parabolic dish colleccor-receiver. This type of solar collector-receiver, when coupled with a heat

uecHMicu 1 .. 1


11/189

engine and generator set, can provide electrical energy for low power applica-tions (kilowatt range). When coupled together in a much larger array theycan produce power in the megawatt range. The NASA through the Jet PropulsionLaboratory (JPL) and the Lewis Research Center (LeRC) has the responsibilityto conduct research and develop the concept for the Division of Solar Technologyof DOE. The LeRC has the responsibility of developing the engines and powerconversion equipment for the point-focus systems.

This study evaluates the concept definition of both free-piston and kine-matic-type Stirling engines for single parabolic dish applications with anominal output of 15 kWe. The Stirling engine system offers the potential forhigh solar energy to electrical power conversion efficiencies and thereforereduced collector size resulting in lower total system costs.

1.2 ObjectivesThe objective of this study program is to develop the concept definition ofboth a free-piston and kinematic-type solar Stirling engine/alternator for asingle parabolic dish application. The study identifies the most attractiveconfigurations for a 15 kWe Stirling engine with an electrical alternatorhaving either three-phase or single-phase i20-208 volts 60 Hertz output. Basedon intended use, the major objective is to develop a Stirling engine/alternatorthat meets above requirements with the highest efficiency since this will re-duce overall system cost. The main goals of this study are as follows:

Parametric analyses of attractive engine/alternator concepts Identification of an advanced solar Stirling engine configura-

tion and its implementation assessment Conceptual design of near term solar Stirling engine/alternator

1.3 ScopeFour major tasks were established to accomplish the stated objectives. The fol-lowing identifies these four tasks and a brief description of the scope of effort:Task 1' Ccnflft^i.oi.Ion uefmition and Parametric Analyses ; Parametrlcallyassess various Stirling engine configurations fur a range of heater and cooler:emperatures over a broad-load profile and identify the most attractive engineconfigurations.

iiexMiicu 12TICMKOlOOtUaCOHPOMTID


12/189

Taak 2: Implamentation Assessmant ; For the one :u ''iguratlon selected fromTask 1, an assessment Is made to determine Its deveU-^pment and productionimplementation potential. This task evaluates prcteut state-of-the-art,produclblllty, durability and growth potential.Task 3; Conceptual Design : In a j-arallel efforc with Task 2, a conceptualdesign for a near-term test bed engine Is established.Task 4; Engine-System Interface : While this study program concerns only theStirling engine/alternator conceptual definitions, the requirements of thereaalnder of the solar power system (i.e., collector, heat receiver, systemcontrols) are provided by NASA to obtain an optimum overall systemconfiguration.

A more complete description of the program scope is contained in Che respectivesections of this report.

1.4 RequirementsThe study requirements are divided into two areas: parametric analyses andconceptual design. It is desirable to have a Stirling engine with a systemefficiency as high as possiblt to reduce collector size and hence overall sys-tem cost. The major program objective, therefore, is to obtain a engine/alternator system with highest possible overall system efficiency whilemeeting all other requirements.

."he program specifications for the Task 1, Parametric Analysis and the Task 3,Conceptual Design are as follows:

Parametric Analyses Conceptual DesignConfiguration:Thermal Input

Heat Metal Temperac-ire:Cooler Water Temperature:Working Fluid:Heat Storage:

Consider Three Concepts Single 15 kW CizeConsider Both Direct

and IndirectEvaluate 650 to 1100*CEvaluate 21 and 65CNot Specified15 Minutes at FullPower

General Electric HeatPipe Receiver Design

SIS'CAS'CHelium70 Minutes at Full PowerWithin Receiver

tfCMNOiOO*weoMourtD1-3


13/189

Parametric Analyses Conceptual DesignEngine Power Modulation: Not Specified Constant Power InputEngine/Alternator Power Various Power Levels 15 kW Constant +Output: 25, 50, 75, and 100 Parasitica

Percent of 15 kW

rieMeioa>


14/189

1.5 Summary of r>e8ult8

1.5.1 Engine Conceptual DeslgaThis study resulted in a conceptual design of a free-piston s.>lar Stirlingengine-alternator which can be designed and developed to meet the require-ments of a near-term solar test bed engine with m^.nimum risks. The conceptualdeslga developed under this study represents an evolution of previous enginedesigns that have been fabricated, assembled and tested by MTI.

The major design changes and significant differences fr^^m past designs are:

1. The heater head represents a novel heat exchanger specificallyintended for sodium vapor condensation heat transfer within aheat pipe heat receiver. The heat exchanger Is designed to beproduced in castings and avoids braze joints which representpotential failure modes.

2. The linear alternator is an inversion of the conventionalalternator design. The stator is inside the reciprocatingmagnetic structure which itself is attached to the powerpiston. This design reduces weight substantially byshrinking the diameter and wall thickness of the pressurevessel.

The conceptual solar engine-alternator desig-. is shown in Figure 1-1. Con-ceptual design parameters are presented in Table 1-1. This engine was cal-culated to have =in overall system efficiency of 38 percent and provide 151cWelectric output. The free-piston engine design incorporates the followingfeatures to provide long life, reliability and maintenance-free operation:

Gas Bearings - Eliminate Wear Close-Clearance Seals - No Lubricant, No Contamination, No

Potential Failure Mechanism Gas Spring - No Mechanical Failure Mode Posted Displacer - Eliminate Need for Close Tolerances

Between Piston and Displacer Internal Supplied Bearings - Eliminate Kxtcrnal Comnrossoi-

^2HH riCHNOvoo'^WHB iHCOiwoMrto


15/189

aIIsisI??

ION

U.QCold Space

Coolant Inlet

:CasC Heater HeadV StufferGas Passage

Annular Heater Space

Piston Gas BearingCheck Valve

/** Piston Gas BearingAlternator Stator

Hoc SpaceDlsplacer

Dlsplacer GasSpring

Displacer GasBearing

Alternator Cooling CoilEngine CoolantDischarge

Fig. 1-1 15 kW Free-Piston Stirling Engine - AlternatorConceptual Design(Dimensions in cm)

791739


16/189

TABLE 1-115 kW FREE-PISTOM SOLAR STIRLIMG EMGIHE COMCEPTUAL DESIOI PARAMETERS

Engine power outputCharge pressureDisplacer amplitudeDisplacer phase angleDisplacer rod diameterCooler tube IDCooler pumping lossAverage cooler delta TRegenerator lengthRegenerate r void vclumeRegenerator effectivenessRegenerator wire a'aiaeterHe^'rer tube numberAverage heater delta THeat from heater to gasOverall engine (before alternator)theraal efficiencyPiston external springDisplacer external springCooler NTUHeater NTUConduction and heat losses

16,500 watts58.2 bars3.828 cm36.1"2.83 cm3.49 mm247 watts27.8*K12.7 cm1308 cm^.991.17 X 10"-^ cm (.003)"4264"C39,760 watts

41.53.1 X 10- N/m^


17/189

ThQ free-piston Stirling engine has no preferred orientation and will operateat any attitude. In addition, the engine interfaces readily with the given GEheat receiver concept as shown in Figure 1-2.

The results of the iBiplesentation study indicate that the near-term con-ceptual design is within the current or is an adaptation of state-of-the-art(Table 1-2). Areas where development is required and, in some cases, isalready in progress at MTI are:

Alternator - Permanent Magnet Controls - Engine-Alternator to Grid

- Engine-Alternator to Receiver Regenerator - Cost Effective Regenerator

Materials Heater Head - Interface with Condensing Sodium

The implementation assessment based on production quantities of 25,000 unitsper year indicated that the engine can be manufactured at a reasonable primecost (direct labor plus material) of $2,500 per engine as shown in Figures1-3 and 1-4. Extrapolation to 100,000 units show a cost of slightly over$2,000 per unit or about $140 per kW. It is clear that the alternator designutilizing samarium cobalt permanent magnetic material represents about 1/2of the total system cost. >fri has explored alternative magnetic materialsthat may substantially reduce en^ine-altr.rnator costs. A high magiietic fluxpermanent magnet ip^terial (Mn-Al-C) has recently been identified as a goodsubstitute for the high cost samarium cobalt. Costing performed for thismaterial is presented in Figure 1-5 and shows costs for the system to be$1300 per unit or less than $90 per kW.

In conclusion, the study developed a design of a free-piston solar Stirlingengine-linear alternator that can provide required power, operate at highoverall system efficiency, have potential for long life and high reliability,and can be produced at a reasonable cost with development effort. It isconsidered thac the frse-piston solar Stirling engine with the above advan-tages will accelerate the commGrclalizatlon of small dispersed i>olar thermalpower systems.

1-8


18/189

SiSliIffHi

Fig. 1-2 Stirling Engine/Heat Receiver Interface791730


19/189

TABLE 1-2STATE-OF-THE-ART EVALUATION

Critical Component1) Heater Head

2) Regenerator

3) Bearing System

Key Technology Condensing Liquid Metal HeatTransfer

Cart Heater Head

High-Volume Processing WithEffective MaterialUtilization

Internally Supplied BearingGas

Surface Coatings Techniques

Technology Status * Significant Improvement

Significant ImprovementFor High Heat Transfer

Adaptation of State-of-the-Art

Adaptation of CurrentTechnology

State-of-the-Art

4) Seals

5) Lisplacer Drive

6) Alternator- Plunger

- Stator

7) Control

Close Tolerance Seal

Posted Displacer and GasSpring

Rare Earth Permanent MagnetManuf acturability

Manufacturing Technique(Mlcrolamination, etc.)

Engine/Alternator StabilityMatching

Displacer Gas Spring VolumeControl

Engine/Receiver InterfaceControl

Extension of State-ofthe-Art for Life

State-of-the-Art





Significant Improvement

* Terins are defined in Section 3.1.2

WfCMANICAirtCHHOlOOi

I-IO


20/189

9alli siill58?

DirectMaterial

V)

Oor-.

V.o


21/189

II

encG Q>*j eI

(U

i5

2500

20004-1nlU0)co

(0Q) -

S*?: 1500t-H

*4-1(0Oo 1000

500

AllOtherEngine

Components

InvestmentCast

HeaterHead

SamariumCobalt

PermanentMagnet

Alternator

\ \ \. \\\\PotentialMaterialCosts^ Increase**\

Potent ial

\forLabor>^;ontent\ \N

>(


22/189

2500r-

I

uout(Qlifl> -c *f>H00 0\C r~.Ui ONOO

'

a,f-Hu H *J


23/189

2.0 CONFIGDRATION DEFINITION STUDIES AMD PARAMETRIC ANALYSES

2.1 Task ObjectiveThe primary objective of this major task was to Identify potential system con-figurations, parametrlcally analyze unique engine concepts, and determine themost attractive configuration for a 15 kWe Stirling engine with an electricalternator having a three-phase 120-208 V 60 Hz output.

2.2 Technical ApproachThe technical approach to accomplish the task objective Is to evaliiate heatInput concepts, determine design points. Identify potential configurations,select three attractive concepts, perform parametric analyses of these con-cepts, and provide a ranking of the attractive confi. iration and recommendations.

To evaluate heat-input concepts, MTI solicited information from Minneapolis-Honeywell in the area of heat collectors/heat receivers and Dynatherm in thearea of heat transport loop. Mutual agreement as to the best concept isdetermined based on the requirements for the Task 1 study; the major influ-ence on choice of heat input concept is system capability for 15 minutethermal storage.

The solar insolation profile provided for the study is reviewed and evaluatedto determine the percentage of peak power (i.e., engine design point) thatwould result in maximum icilowatt-hour output over an annual period.

The concept definitions are based on past experience regarding engine-alternatorsystems that have the potential for maxirauii. overall system efficiency. Beforeinitiation of the parametric analyses, the configurations identified arereduced to three (via an initial ranking of configurations). The parametricanalysis is performed to provide the necessary perfOiTmancc data of eachconcept in comparison with others such that each configuration rpuld beranked and a recommendation made.

UCCmAWCM a .r>CHHOioa> ^JLINCOaWOMTtO


24/189

2 .3 Heat Input Concepts

2.3.1 An Integrated Heat Receiver, Thermal Storage and Heat TransportSystem Concept

The objective of this subtask was to conceive a heat input concept which willtransfer the heat from the Cw*llector to the Stirling engine working fluidaaiifmlzing any penalty on the engine's overall thermal efficiency. Theheacer head configuration of a Stirling engine is designed to accomplish itsthermodynamic function of adding adequate heat to the working fluid with min-imum dead volume and minimum pumping losses. Any deviation from the optimumconfiguration will result in degradation of engine performance. Therefore,the ideal heat receiving and heat transport system operates in such a mannerthat the optimum heater head configuration is maintained.

A major requirement of the system for Task I study was the capability of thethermal storage system to provide 13 kW output for 15 minutes. The thermalstorage can be either built as an integral part of the engine /heac receiversystem or at a remote location. Because of the high operating temperatures(around STl'C to 927''C) a remote heat storage will introduce additionallosses and system complexity. Also in a system configuration where thi thermalstorage is remote and fixed, and the engine is mounted on the collector, flex-ible joints and liquid metal pumps at high temperatures would be required.

Sased upon the aforementioned considerations and initial assessments, effortswere directed at conceiving a system which integrates heat receiving, heattransport and thermal storage as a single unit, as well as being sufficientlycompact to be mounted on the collector. No further work was performed on theremote heat storage system because of anticipated losses, complexity and non-availibillty of liquid metal pump in desired temperature range.

2.3.2 Heat Input Concept SelectionSeveral heat input concepts were examined and identified as follows. Theinitial assessments of advantages/disadvantages were reviewed with Minneapolis-Honeywell and Dynatherm in an effort to identify the most attractive heat in-put concept. The heat input concept selected is a liquid metal system (essen-tially a sodium pool boiler) using a molten salt for heat storage. The systemis compact, lightweight and operates without liquid metal pump. A summaryof concepts considered with advantages/disadvantages leading to the selectedconcept are listed on the following pages.

ticMMoiooT 2-2mcMMaitno


25/189

DIRECT RADIATION

RADIATION - CAVITY RECEIVER

A. External AbsorberThis idea was abandoned due to the followingdisadvantages : Small heat storage capacity High heat losses Uneven heat flux

TO ENGINE Reradlation to nearby structures Secondary heat transport medium required Heater head would require modification

B. Internal Absorber1. Direct Radiation: this idea was not pur-

sued due to the following disadvantages: small heat storage uneven heat flux hard to control heat flux more complex heater tube geometry longer heater tube - less engine

efficiency2. Cavity Receiver: this concept was rejected

because of the following disadvantages:

HEATER TUBE

HEATER TUBE

y3:>~_c:

low storage heat capacity to weight ratio less efficient utilization of reradla-

tion-tube area comparable to aperaturearea

increased heater tube area required3. Liquid Immersed System: this system has the

following advantages over the foregoing sys-tems as follows: nearly uniform temperature high-heat transfer race higher heat storage capacity (higher

liquid specific heat Chan metal)

mctuMicMriCMNOlSO'McoMOwreo 2-3


26/189

natural convection possible, no pumps requiredNo further considerations were given to the concept, however,due to the following disadvantages; low heat storage compared with heat of fusion or heat

of vaporization system system temperature drops as heat Is abstracted, resulting

In low engine efficiency large mass required to maintain allowable temperature

drop large mass and heavy system result in more heat loss

areaC. Two~Phase Heat Input System

1. Liquid Vapor System: utilization of latent heat of vaporizationas heat storage.

Several liquid metals have properties that are ideal for thermalstorage such as: high latent heat of vaporization low vapor pressure high thermal conductivity high specific heatAdvantages of this system are: high heat transfer rate, optimum heater head configuration high critical heat flux for boiling, no hot spot in cavity temperature modulation possible power drain regulation possibleSystem disadvantages are: use of vapor-state results in poor volume to heat storage

capability ratio use of liquid state results in poor mass-to-heat storage

capability ratio

MieMMICU 24nexmKoofiNCOnfOM'ID


27/189

It is apparent that the liquid-vapor system is very attractive from


28/189

Iill

EXPANSION VALVE

Fig. 2-1 Two-Phase Two-Teraperature Sodium System VERTICAL792437


29/189

2. Solid-Liquid System or Molten Salt System: many salts havethe properties that are ideal or thermal storage such as: high melting point for high temperature application high heat of fusion high specific heatThe disadvantages of such a system are: low thermal conductivity of salt; heat of fusion utiliza-

tion requires a large heat transfer surface a second high-heat transfer working medium is reouired9 complex and costly system as shown in the following

sketches

W0I7KINGFLUID INLET

These disadvantages were overcome, and the related problemswere solved leading to the pursuance of Che reconunended concept.

2.'}.. 3 Recommended Heat Input ConceptThe ir.olte-.i salt system has a high heat storage density. It also has chesame advantages as the liquid-vapor system if liquid metal Is used for the heat

WCMIMCUrtCMMvooTINCOIIMIIATtD

2-7


30/189

transport ondlum. The main disadvantage of this system, as discussed pre-viously, is the low theniL.1 conductivity of the salt. The ability to transferheat into or out of the salt requires that a large heat transfer area has tobe provided. To increase this heat transfer area, the salt is distributed ina large number of small thln-walled containers, usually small diameter cylin-ders. The numerous amount of cylinders makes this system complex and costly.

2.3.3.1 General Description of Recommended Concept. For the anticipated wideapplication of solar power systems, it is clear that the cost of each componentis an Important consideration. To make the molten salt system siiiq>le andeconomical, a method has been conceived whereby the system structure Is simpli-fied. Instead of using individual small-diameter cydindrical containers,the proposed system utilized a single, contlniious tube coiled in an enclosureas shown in Figure 2-2. An enclosed 2.54 cm outer diameter (OD) by 2.22 cm

2inner diameter (ID) tube, 51.8 maters in length provides 4.13 m outer surfacearea for heat transfer. The heat of fusion of 35.8 kg of LIF alone Isadequate to rtm the engine-alternator for 15 mlnu.ss at IS ktf full power output.

Liquid sodium operating at STI^C, 28C above tae melting point of LiF, wasselected as Che heat transport nsedium. The vapor pressure of sodium at 1600Fis 13.28 psla and the net pressure acting on the system is only 1.42 psi.This low pressure makes a thin wall structure poti^lble. As a result, thetemperature drop across the wall Is low for the entire heat transfer surfacethereby allowing a higher engine operating temperature and leus thermal stressfor the same receiver material temperature limitation. This system is compact, lightv^ight and economical and operates without a liquid metal pump andhigh receiver system efficiency. The system specifications are listed inTable 2-1.

2.3.3.2 Receiver System Heat Losses, lae heat loss of the receiver systemwas calculated and results are as fellows:

a. Heat Loss During Steady-State Operation;The total heat losses during steady-state operation is4.2 kW.


31/189

9Bliill

to

14.2 cm.HEAT INPUT

LIP STORAGE

aoooooooooooooa 3DD, az;.zo

903*C \ 87rcSURFACE TEMP SODIUM HEAT-4- TRANSFER

SOLARCAVITYRECEIVER

ooooao ::

LIQUID60 KGOOO

oz:;;zDozz::zx)Doooooooooo

TUBES

STIRLINGENGINE ENVELOPE

Fig. 2-2 Recommended Concept792430


32/189

TABLE 2-1SPECIFICATION OF A LiP-Sa SYSTEM

LIF WeightLIF Volume at Liquid StateContainer TubeTube Surface AreaTube WeightSodium WeightOperating TemperatureHeat Flux at CavityTemperature Drop Across Cavity WallInsulationReceiver Efficiei.cy

35.8 kg.02 m^2.54 cm OD x 2.22 cm ID x 51.8 m long4.13 m^48 kg75.3 kg871C1b9,779 w/m231.8C10 cm Ktiowoll 12 lb density91 percent

UtCMANICAiriCMNOtOaTiNfOlirOMTIO

2-10


33/189

b. Heat Loss During Shutdown Period ;

During the shutdown period, no heat Is added to the system butthe heat loss persists as long as the system temperature Ishigher than the ambient temperature. To ensure chat the sodiumIn the receiver system will not solidify (for a reasonable timeperiod), system temperature as a function of time was calculatedand plotted as shown in Figure 2-3.

2.3.3.3 System Efficiency .a. Collector Size :

The collector size was determined as 8.2 meters in diameter, basedon the maximum alternator output of 15 kW at maximum solzr in-solation and the following parameters: Maximum solar insolation 1.03 kW/m

2 Receiver shadowing area .557 m Stirling engine overall themal efficiency (assumed) 40% Alternator efficiency (assumed) 90% Syttem operating ten-porature STl'C Total receiver system heat loss 4.2 kW Collector reflectance S'%

b. System Efficiency :With the available heat storage, the engine operates at a constantdesign temperature during cloud covering periods. The heat lossremains constant due to constaiu system temperature. As thesolar insolation decreases, the heat input decreases but theheat loss is constant. Systjm efficiency therefore decreases asthe insolation decreases. The system efficiency as a function ofs^lar insolation was calculated and is shown in Figure 2-4.

2.3.3.4 System Performance . Using the June 21, 1976 solar insolation profilein Lanchester, California as an example, the receiver temperature variation,power output, insolation outer surface temperature and receiver system heat

ICMAWCIIl 211TtcmwiOOTMCODKMUTIO


34/189

>F C

HI

UJoc

n: (jj I-

UJ

1700

1600

1500

1400.

900*

V. RECEIVER TEMP.SCAtJ >

-800'

OUTER WALL TEMP^

"^0 =700-1200 nnn ' temp, of receiver temp, isocf prior-600 to shut off

2. WIND VELOCnv 15 MILES/HR3. INSOLATION: KNONOLL 4"4. APERATURE COVERED WITH 4" INSOLATIONy^jQ DURING SHUT DOWN PERIOD - NO RADIATION LOSS

inoo -600

800 00700 600

1 10 11 1?f'oiiR:

Fig. 2-3 Received Temperature VersusHours After Shutdown

13

40**110

100

9030"-

]" 15

UJ

aeI-o

792439


35/189

1. COLLECTOF: REFLECTANCE; 85';;TRACKING ERROR IGNORED

2. RECEIVER LOSSES4" KNOWALL INSOLATION 12 LBS DENSITY15 MILE/HR WINDAMBIENT TEMPERATURE 70"F

IOC

80

60>-

RECEIVER EFF

COLLECTOR RECEIVER SYSTEM EFF.

SOLAR INSOLATION kW/M'

WCW OMTIDFig. 2-4 Collector-Receiver Efficiency

Versus Solar Insolation2-13 752440


36/189

loss were calculated and are plocted versus time in hours as shown In Figure2-5.

2A Engine Design Point SelectionThe englne-alr


37/189

dill


38/189

level and is not directly available but can be obtained by utilizing the plotof insolation flux versus average hours per day for each month as shown inFigure 2-6. For the purpose of analysis, the insolation flux density was

2 2divided into ten levels between 0.5 kW/m and 1.0 kW/m by an Increment of2 2 2.OS kU/m and three levels between .2 kW/m and .5 kW/m with an increment of2 20.1 kW/m plus those above 1 kW/m .

At each insolation level, the average hours per day were scaled from the plotand then multiplied by the numbe ; of days in that month. For example, in the

2 2month of March, the total hours at insolation between .9 kW/m and .95 kW/mis 1.64 hours x 31 days 50.84 hours, as shown in Figure 2-6. This processwas repeated for the fourteen insolation levels for twelve months and the re-sults are summarized in Table 2-2 . With the annual total hours av. each insola-tion level so calculated, the annual kW hours available to the collector pereach square meter of the collector area is obta-'ned by dimply multiplying thetotal annual hours by the average insolation in chat range. For example, inthe month of March available kW-hr to the collector at Insolation levels be-tween .9 kW/m^ and .95 kW/m^ is 563.52 hours x '^ ^ '^^ = 521.25. The availableenergy to the receiver is reduced to 441.06, which is 85 percent of 521.25,representing reflectance of the collector.

The operating temperature of the recommended receiver system is fixed at 1600*'F(as well as the heat Iosf; by adjusting the power output level according to

2the solar insolation level. For unit consistency in Table 2-2, kW hr/m ofcollector area, the heat loss due to radiation and err 'ection at the receiver

2is converted into loss of energy in kW hr/m at the collector area. Th .oUv.l2receiver heat loss is 4.202 kW-hr/hr equivalen*- to +.202 kW-hr/hr * 52.34 m

or .08028 kW-hr ar u; at the collector independent of insolation level. Inthe above example, the 443.06 kW-hr available at the receiver are collected in563.52 hours. Therefore, the energy available to the engine in the month ofMarch is 443.06 kW-hr/m^ minus (563.12 hours x .08028 kW hr/hr m ) - 397.83 kW

2hr/m . As previously stated, th-i engine alternator system is designed todevelop 15 kW output at maximum solar insola.,ion. At each lower insolationlevel, the alternator output is listed in Table 2-2.

i((e*uMiei 1 iNeoiiai


39/189

1200


40/189

TABLE 2-2ANNUAL DIRECT INSOLATION IN LANCASTER. CALIFORNTA

I00

~.^,^80ltlOO


41/189

Engine Design Point Selecti on. The Information listed In Table 2-2 wasplotted In Figure 2-7 which shows annual hour distribution and energy availableto the engine versus Insolation flux. The engine design point was selectedat the peak of energy avall.-ible to the engine, rather than the nour distribu-tion peak. This Is because the quantity to be maximized is the annual energyproduction. At the peak of the energy curve, the engine output is 14.1/. 9 or15.7 kW (alternator efficiency of .9 assumed). Therefore the engine efficiencymust be maximized at IS. 7 kW level to obtain the maximum kilowatt-hour over anannual period.

It Is also Important to note In Figure 2-7 that the 90 percent of the totalenergy available to the engine occurs at powe^- levels greater than 30 percentof the peak Insolutlon. This implies therefore that the Stirling engine onlyrequires power modulation over the range of 50-100 percent power to utilize90 percent of energy available.

2.5 Concepts Selected Tor Parametric Analysis


42/189

800

700

600

500 -

I I3 ZoC ID< aa_i >- e/15C a>-o toQ Of

400

^g 300 I"

200

ICO

FMniNE OESinM POIfiT0

Al TERNATOR OUTPUT kW

ENER'^YAVAILABLE TOTHE ENGINE

I X JL . X.1 .2 .3 .4. .5 .6 .7 .8

ANNUAL DIRECT INSOLATION FLUXFig. 2-7 Energy Available to the Engine

MoowMATio versus Insolation Flux2-20

.9 1.0

16

15

14

n\2

\^

10

4

3

o.IoR o

a:

792M2


43/189

SIBliHiw

roIS4

Teflon

*Powr

Piston/AlternatorPlunger

^/^*>'>**

Kegenerator

'

Heater Tubes J

Fig, 2-b b kW hnglne Utilizing Wear Seals(dimensions in cm)

Displ.tcer C.as BetnrivOisplacer

;i;'S


44/189

three alternators are required for three-phase output. The engine embodiedthe following design features:

Rubbing contact filled Teflon bearings and seals on the plunger Gas bearings and close clearance seals on the dlsplacer Tubular heater head Gas springs to tune both dlsplacer and plunger Barium ferrite permanent magnet alternator plunger External stator alternator

Resulting from these features, the design has certain inherent advantages anddisadvantages

Advantages Contact seals and bearings are simple, low cost and adjustable Tubular heater head leads to efficient heat transfer and low

mass Highly reliable gas springst Seals and bearings easily adjustable and replacoabl-j Alternator of low co:;t design

Disadvantages Performance deterioration due to seal wear Starting friction due to rubbing contact seals Bearing alignment problems due co locating rear bearing in

pressure vessel Contamination of gas bearing and regenerator due co west

particles from tarings and seals Difficulty in Jtctaininj- sufficient pr.ston gas spring due to

large interna! volume of alternator Excessive weight d(ie to pressure vessel size

2-22coMoiunD


45/189

2.5.1.2 One Fifteen kW Single-Cylinder Engine with Dual Gas Springs . Afifteen kilowatt engine has the same general characteristics as the five kUengine except scaled to a higher power. This engine was initially envisionedfor use with an external phase converter to supply the specified three-phaseoutput

No additional advantages would appear to result except perhaps some slightlysmaller thermal losses to the surroundings due to the higher volume to surfaceratio of the engine.

Additional disadvantages would be a more severe gas spring problem caused bythe larger ratio of alternator volume to engine displacement, and higherspecific weight. (A larger, thicker pressure vessel is required because poweroutput from the alternator is essentially a surface function while engine out-put is essentially a volume function causing alternator size to grow at a morerapid rate than the engine.)

2.5.1.3 One Fifteen kW Single-Cylinder Engine with Mechanical Spring . Thisis an engine with the same general characteristics of the previous engineexcept it utilizes a gas bearing on the piston and a mechanical spring onthe power piston/plunger. The mechanical spring turned out to be quite largeand one-third of its mass must be added to the moving mass of the plunger.Also, any asymmetry in the spring or its mounting surfaces could impose sideloads on the piston gas bearing that are large enough to exceed a reasonabledesign capacity. Experience with other high-cycle springs has indicatedthat the necessary material and manufactuting standards are so high thay theypreclude the manufacture at reasonable cost.

2.5.1.4 One Fifteen kW Torsional Engine . A 15-kilowatt engine with two parallelcylinders and a torsional oscillating alternator coupling, with two rocker-iike pistons, is shown in Figure 2-9. It was thought that this confipurationmight simplify the gas bearing problem since there was no gas spring per se.Expansion in one engine resulted in compression in the other engine. Thusthere faere no gas spring losses and the alternator chamber had a constantvolume. Also, the effective moving mass would be lower since in rotary motiononly a portion of the mass moves at the full velocity, while mass elements

_ 2-23ftCMHOtOOTNCOHMIMTtO


46/189

Fig. 2-9 15 kW Torsional Engine

IKOIWOMrfO 2-247977'


47/189

closer to the centerllne move slower and possess less kinetic energy. How-ever, the mechanical losses from the large number of required bearings seemedas though they would outweigh any gains made by eliminating the gas spring.Consequently, the concept was rejected.

2.5.2 General Trade-offsThe necessary trade-off/review of the general concepts presented in Subsection2.5.1 and the configurations finally selected for the parametric analysis arediscussed herein.

2.5.2.1 Direct versus Indirect Coupled Alternators . Consideration was givento various means for coupling the free-piston Stirling engine (FPSE) to theelectrical generating means. The basic decision was first between directcoupled (i.e., FPSE (Figure 2-10) mechanically attached to an appropriateelectrical machine) and indltect coupling. Indirect coupling, e.g., hydraulicor pneumatic, would allow rotating electrical machinery to be powered by thevariable stroke free-piston Stirling engine (Figure 2-11).

Supplementary benefit could be derived fron the engine concept shown in Figure2-11 the area of frequency control. Since the generating equipment could beinduction type, the frequency would be solely dependent on the bus the unitis attached to and three-phase output is thereby possible.

However, the indirect coupled systems suffer from a double-ef ficiancy penaltysince the work done by the engine must first be transformed from a linear toa pressure function, then retransformed to a rotational system. The inherentloss at each transformation indicates that the number of transformations shouldbe held to a minimum. Consequently, it was decided to pursue concepts usingdirect coupled engine-alternators.

2.5.2.2 Gas Bearing versus Wear Seals . Attempts were also made to incorporatehigher efficiency subsystems. Gas bearings were selected over rubbing contactsystems for lower parasitic losses, lower wear rates, less internal enginecontaadnation from wear particles and higher inherent reliability. It isalso possible to externally pressurize the engine gas bearings before theengine is started to reduce bearing wear and facilitate startup.

2-25MCOH'OMTtO


48/189

H" ter Tubes

Regenerator

Cooler

"ower Cylinder

Alternator

Fig. 2-10 Three i'ree-piston Displacer Engines Driving ThreeSingle-phase Linear Alternators

tncMwicMriCMMOiaOTINCOIIr>OIITtO

2-26


49/189

Heater Tubes

Displacer Piston

Regenerator

Cooler

Power Piston

Pump Piston

LAlternator

Diffuser

Fig. 2-11 Linear Engine with Fluid Dri^'c

HtCNANtC-'*!ncNMOiootiHco>ourto

2-27


50/189

2.5.2.3 Mechanical Spring versus Gas Spring , The design of a mechanicalspring was Invescigaced in order to take advantage of its potentially higherefficiency. However, a detailed design effort showed serious problems ofhigh-cycle fatigue would seriously impair engine life. Furthermore, one-third the weight of the mechanical spring contributes to the moving mass andthis, in turn, raises the bearing load. Additionally, any nonsymmecry inmanufacturing the spring or its mounting surfaces will result in sizable sideforces applied to the piston bearing due to the very large axial forces re-quired. Because of the small magnitude of gas spring losses, it was decidedCO restrict the proposed couflguratloas for parametric analysis to those whichincorporated gas springs and close clearance seals. The magnitude of the per-formance gain utilizing mechanical spring would not justify the sacrifice inreliability.

2.5.2.4 Inside versus Outside Alternator Stator . The alternator configurationwas also subject to some scrutiny before a particular approach was settledupon. Detailed discussion of linear alternator operations and trade-offparameters are presented in Section 2.6. The early concepts involved theso-called external stacor construction where alternator stator surroundedthe plunger and the moving magnets were mounted on the plunger outside surface.The requirement for gas bearings and close clearance seals meaft that therewould be corsiderable length consi'med by those systems. The pole arearequired for the permanenc magnets also required additional length.

The inside of the plunger was hollow and essentially wasted space. Also, theexcessive length of both the plunger and stator Ted to manufacturing and op-erating difficulties. Internal diameters with length/diameter ratio of morethan three become very difficult to grind in a single operation withoutspecial tooling. Furthermore, the plunger is difficult to hold straight andconcentric from end-to-end.

The alternator was then redesigned with the thought o reducing the complexityand relaxing production tolerances. It was noticed that if higher fluxdensity magnetic materials such as samarium-cobalt were used for the movingmagnets, the pole area could be reduced so it could fit within the bore ofthe plunger. This led to the creation of the concept of internally mountedmoving magnet rings coupled with a cantilevcred inter stator.

*^ HtCHAHlCM n mATtCMHMOO? Zi.Hmcemromtno


51/189

When chls configuration was adopced, other design simplifications followed.The gas bearings and seals could be one continuous diameter and "telescoped"over the alternator thereby allowing the cylinder to be shortened to providea length to diameter ratio of about three; it could also be manufactured inone piece, eliminating alignment problems. Because of the continuous diameter,the piston/plunger could also be constructed and machined as a single piece.The high flux density magnetic material which allowed the compact constructionalso has a very high resistance to magnetization, making it virtually impossi-ble to demagnetize the alternator during operation. This is particularilyimportant during the development stages when many unpredictable situationsoften arise.

Other benefits came from the Internal stator too. The piston/plunger weightwas reduced to a minimum since the material was "used twice". The steelpiston also provides the return flux path for the magnetic circuit, makingadditional back Iron unnecessary. Finally since the alternator is so compact.It is possible to use a much smaller, lighter pressure vessel to contain it,and the internal volume of the vessel is closer to the desired optimum foreffective gas spring action.

2.5.2.5 Cast versus Wrout^ht Heate.- Head . The cylinder head construction wasalso suoject to considerable thought. Because of Che decision to use indirectheating of the engine heater head, coupling it to the receiver with a heatpipe and thermal energy storage module, it was felt that the number of jointswithi.i the heat ^ipe envelope should be minimized. This decision was based onthe extremely active nature of hot sodium. Weld joints, braze joints andinterfaces of dissimilar metals are all subject to possibly serious corrosion.Consequently, an attempt was made to minimize the number of joints, and limitthose absolutely necessary to nonstructural areas. In this manner i. was con-sidered that the joint designs could be very conservative and the chances offailure would be minimized. For the above reasons a cast heater head wouldbe desirable, since the entire surface Inside the heat pipe would be of asingle material, the only Joint being at the point where the head joins theheat pipe. To guarantee reliability, a redundant joint design could be usedat that location.

mouMicu 2~*29rtCMMKOOIiNCOWOMnO


52/189

Ceramic heater heads were considered, but structural ceramics compatible withhigh temperature Imodium are not readily available at this time. Other diffi-culties result from the difference in thermal expansion between ceramic andmetal parts, but many of these can be overcome if joint design is done withcare. Additionally, u. e principal advantage of ceramic, i.e., high tempera-ture strength, is of doubtful value since the increased radiation losses fromthe receiver may wipe out any gains in engine cycle efficiency. However, aceramic heater was included in the parametric analysis so that the effect onefficiency could be assessed quantitatively.

2.5.3 Layout/Description of Selected ConceptsThree (3) engines were defined for parametric analysis:

1) 3-5 kW engines with gas bearings, close clearance seals2) 1-15 kU engine-alternator with gas bearings, close clearance seals3) 1-15 kW - same as the above, except incorporation of the

ceramic heater head

The first engine is a 5 kilowatt design (Figure 2-12) using a cast cylinderhead with internal slots to serve as heater tubes. The head is then linedwith a "staffer" which closes the slots and forms passages which travel froma small port in the center of the head radially, then axially back to thetop end of the regenerator. The regenerator is annular, packed with layersof 200-mesh screen woven from .001 in. diameter wire. The cooler is definedas a number of round passages parallel to the engine axis, allowing the ac-tual hardware to be either fabricated from tubes or drilled from solid. Thedisplacer and power piston are both carried on gas bearings, using closeclearance seals. The displacer and power piston are tuned using gas springs.The large bore/stroke ratio in this engine, along with the volume within theplunger occupied by the alternator stator, allows the entire pressure vesselvolume to function dS the gas spring. Because of the pressure variation inthe whole cavity a plenum must be provided for both the high and low sidesof the gas bearing system. The gas bearing supply is ported from the pistongas spring.

MfChamCM 230


53/189

I,%n2*

iLpJ

in

Coolant PassageCas BearingSupply PlenumCas Bearing

Drain Plenua

Alternator StatorPlunger Magnet Piig

Power Piston.Displacer Gas Spiiag Dlsplacer

CenteringPi enum

80

Fig. 2-12 Five kW Engine-Alternator(Dimensions in cm)

791736


54/189

The alternator Is a permanent magnet Internal stator design of the type de-scribed in Section 2.5.2.4 and is cantllevered from the end of the cylinder.

The second configuration (Figure 2-13) was schematically the same as the 5 kWdesign, but with the power scaled up :o 13 kW. The third design was considereddimensionally the same as the 15 kW engine with the assumption of a ceramicheater head and any other parts required to raise the temperature limit of theengine.

2.6 Linear AlternatorsThe free-piston Stirling engine produces a reciprocating motion. An energyconverter is required which will convert the mechanical energy in this re-ciprocating motion into electric energy. Such an energy converter, called alinear alternator, typically has one member directly connected to the pistonof the free-piston Stirling engine and the other member is stationary. Anyelectric energy converter obeys the principle of reel- rocity, i.e., it con-verts mechanical energy into electric energy and vice versa. Various formsof linear electric motors can therefore be used as linear alternators.

The principle of operation of these various motors can be diviaed info two maincategories: flux switching and flux reversing. In the flux switching type,the DC excitation coil, which is the source of the main flux, is on the sta-tionary member as are the AC coils. The flux linkage of the AC coils ischanged (switched) from high to low level by the movc:ment of the plunger be-tween its two extreme positions. In the flux reversing type, the source ofmagnetomotive force (MMF) is on the moving member or plunger. The coils areon the fixed member or stator and the flux linking then goes through apositive maximum to a negative maximum as the plunger moves between itsextreme positions. 7i. either case, the change in the flux seen by the ACcoLls causes a voJtage to be induced in the coils of the same frequency asthe oscillation o.' the plunger. When used in the alternator mode, this in-duced volta'^*- del^''ers current and power to the outside load. The source ofthe excitxng flu:: (MMF) may be a coil; i.e., electromagnetic or it may bea permanent magnet. Use of a permanent magnet for flux reversal machinesavoids sliding or flexible current colle-itors.

ticiwetoot i.~ ii.WCOWOMnB


55/189

s,

reIU9

Heater Head-Stuffer

^

Coolant PassageCooler Tube

Gas bearingSupply Plenum

EzzTTT

Gas BearingDrain Plenun

zr'*i'vr'

heaterTube

DispluierCenteringPlenum

Plunper Mat^nt-t Ring

^Mteniator Stator

94.'9I7J7

Fig. 2-13 15 kW Preliminary Conceptual De.-lgr.(Dimensions in cm)


56/189

Variants of these motors can be classified from yet another viewpoint, depend-ing on whether or not the moving member is used as a flux carrier. In eitherthe flux stritchlng or reversing type only a part of the W^etlc circuit needst^ move relative to the other part. la some configurations this is achievedand result's in aa extremely light plunger. However, in such configurationsthe flux has -.o cross twice as maay air gaps as required in the configurationswhere the flux carrier also travels with the mcving part of the magnetic cir-cuit. In the flux switching type, there are two modes of operation, namely:saturated and unsaturated plunger.

Ttie following characteristics are common to all these alternators: Osclllacicg plunger Stationary stator No moving contacts anywhere Single-phase AC power

From a systems viewpoint, the following are desired characteristics: Low plunger weight Low side pull to reduce requirements in bearing design High efficiency Low cost Ability to scale to high powers, enc.

2.6.1 Selection of Types of AlternatorsTo T^ake a comparison between various types of alternators, it is desirable jodesign each concept to a given specification of power, weight of plunger andefficiency, etc. and then evaluate the competing designs. The choice is madeon ^he briis of t.. e qualitative judgements with a reasonable certainty ofhav'ng selected the best alternator type for the given application. Table 2-3-summarizes the characteristics of these various types. Some of the properties,like the magnitude of the side pull, etc., are inherent in the concept. Certainethers, like efficiency and weight of the mo\:ng plunger, are design relatedand dependent on other considerations. For example, one could design to obtainsimilar efficiency in the various concepts, but the weight of the overallsystem in these different concepts will be considerably different from eachother. The qualitative statements used in Table 2-3 approximately representthe design of any particular concept.

2-34


57/189

iIf

I

Table 2-iComparlaon of Three Different Concepts of Linear Oaclllating Machines

Flux Switching

Unsaturated Plunger

Flux Variation Maximum to low(does not change sign)

Total Weight

Plunger Weight

Side Load

Efficiency

Ability to scalef-o lilgher powers

Control

Saturated Plunger

Maximum to low(does not change sign)

High

Medium

By increasingdiameter only

Four control variables1) Ki angle i i) stroke

ill) D.C. excitationiv) applied voltage

Very low (if movingmember is nor used asa flux carrier)

Very low

Medium

By adding additionalsections

Four control variablec1) a angle 11) stroke

111) D.C. excitationiv) applied voltage

Flux Reve


58/189

2.6.2 Reaaons for Selecting Flux Reversing Type Permanent Magnet AlternatorThe permanent-magnet reversing flux-type alternator was selected* based onthe previous considerations, for further study. It offers the followingadvantages over the other types considered:

Light-plunger weight Potential for high efficiency Excellent geooietrical match with the engine. (The diameter of

the alternator plunger and the diameter of the piston of theengine can be the same which results in considerable simplifi-cations.)

Overall light weight of the alternator Low side pull making gas bearing design easier.

The lack of one of the variables for control (viz., the DC excitation) isreally the only penalty associated with this choice.

2.6.3 Two Configurations of Permanent Magnet A-ternatorThe permanent-magnet reversing flux-type alternator can be implemented in manyconfigurations. A configuration with the plunger in the form of a hollowcylinde'- lends itself to interfacing with the engine as shown in Figure 2-14.Here the nollow space inside the plunger can be used for engine piston, gasspring, etc.

From the heat transfer consideration, the outside stator arrangement of Fig-ure 2-14 is desirable since almost all of the heat is generated in the stator.Magnetically, the design of the stator is straightforward. The highest fluxdensities occur at the airgap (i.e., at the plunger diameter). This allowsthe use of flat laminations.

An 'inside stator' configuration was also considered since it offers the fol-lowing advantages (refer to Figure 2-15)

1. Smaller overall diameter for the same plunger diameter (and,hence power)

2-36


59/189

9Mlill

STATOR LAMINATION

X' / } ? I t I ^y'/ V I / t t /zbsI i . I r N I I a. I >~7CTr^

CYLINDRICALMAGNET

DOUGHNUTSHAPEDCOILAIR GAP

6ACK IRON SLEEVE

N iT s I 1 N 1 I 8 ] I N rrsSTATORLAMINATION

Fig. 2-lA External Stator Configuration''))6e3


60/189

SIfill

o *?BACK IRON

I

00

CYLINDRICALMAGNET STATORLAMINATION

PTFP V 7 r ^ V ni'i' ^i/r ^r^ri AIR GAPDOUGHNUTSHAPEDCOIL

no. rjTTi .rTTD rrxryni^TUj;

BACK IRON FOURETURN FLUX

CVLIN&mCALMAMCT

AIR dAP

STATORLAMINATION

Fig. 2-15 Inside Stator Configuration791682


61/189

2. Sfoallec overall weight3. Smaller steel welghc Implyi % less core losses4. Smaller diamecer resulclng in Che smaller mean diameter of the

copper wires producing smaller resistance. (The resistancecould be as much as 30 percent lower than the outside statorconfiguration for otherwise identical alternators.)

5. Considerations (3) and (4) imply the potential for hipherefficiency.

The difficulties associated with this design are:1. Heat transfer area is small and, hence, alternator cooling im-

poses more difficult requirements on the cooling system.2. The stator has to be cantilnvered from one end. This would re-

quire higher clearance between the plunger and the stator.Furthermore, this limits the size of the alternator for a givenplunger diameter.

2.6.4 Parameters of DesignThe airgap power in a permanent magnet alternator is given as follows:

P .(4.8 X 10^) (. D "r W)b2 X (|) ji - -^ j'^' [T^.] 1ivnr \ -" -- ^'-V^> C'where

P = Airgap power (watts)D" = Diameter (in inches)1" = Half length of each section (inches)1" = Amplitude of maximum strokeIm" = Thickness of the magnet (inches)g" = Radial length of the airgap (inches)Br = Residual flux density of the magnet (Tesia)(V/E) = Ratio of applied voltage to induced voltageE = R.M.S. value of the fundamental component of the induced voltage

with slot opening consideredE = R.M.S. value of the induced voltage with no slot opening

fOwaiieM 2-39ncMnoiooTmewtfOiuTto


62/189

then

If

f, = E/E^S /i- Racio of slot opening to the length of each section

X = Airgap reactive of the machine with slot opening and fringingflux considered

X - Airgap reactive of the machine with no slot opening and assunt-ing the AC coil current flux to have only the radial component

thenC Xf Xo

The above equation takes into consideration the stability requirements dis-cussed in detail in Subsection 4.7.1. Figure 2-16 presents the permanentmagnet aZternator nomenclature.

2.6.A.1 Selection of the Magnet Material . The magnet properties of Lmporranceare shown below:

Br, residual flux density He, coercive force y, recoil perraeability iD , resitivity d, density.

The previous equation shows that for a given magnet volume (2 tt d" 1" Im") thepower output increases as Br". However, for stability reasons, V/E has to begreater than unity, and this can also be seen from the reference to the abovepower equation. Thus, if the plunger is not moving and the alternator is con-nected to the power grid, the current in the AC coil will cause a reverse fluxin the magnet. It is important that the magnet should not demagnetize undersuch conditions. This implies that 'He' the coercive force of the magnet,should be high. Altema: '.vely, the rated voltage siiould 02 limited to such avalue that the flux due to current in the AC coils will not demagnetize themagnet. Thus, higher values of coerciv*- torce will result in higher powerdensity.

MCHaMCMTtCMMtaOTHKOIWOIUto 2 M


63/189

s.II?

I

Permanent Magnet7- ////// ^ / /y// ///////////

s N

tip

K t, -H;r

U2s;;-j

w"^r

Statop

2 I-

Centerline


64/189

As the plunger oscillates back and forth, the flux density at its surfacevaries due to the stator currents and also the permeance variations cruisedby the effects of slot openings. This flux variation will cause surfacelosses due to the eddy currents in the surface of the magnet. To keep surfacelosses to a minimum it is then desirable to have as high a resistivity aspossible. The density of the magnet material should be as low as possible toobtain the lightest possible plunger weight.

Other physical properties such as temperature stability, coefficient of thermalexpansion, etc. also have to be considered during the design phase.

Properties of some commercially available magnets are listed in Table 2-4.

Based on these properties a "Samarium-Cobalt" (SruCoe) material was selectedfor the designs to be parametrically analyzed.

2.7 Parametric Analysis

2.7.1 MethodThe receiver system characteristic has been discussed in Subsection 2.3. Theengine power level at which the maximum thermal efficiency should be optimizedhas also been determined in Subsection 2.4. With a number of physical con-straints, such as material property limitation and alternator requirements,a parametric analysis for the optimization of the engine under differentoperating conditions was performed. The objective cf this analysis is todetermine the operating condition at which the highest possible thermal effic-iency can be achieved. The analysis consists of ten operating conditions,five heater-head temperatures (1200, 1400, 1600, 1800, and 2000F) and twocoolant water temperatures (70 and 150*?). After tlie operating condition forthe highest thermal efficiency has been selected, the same engine i? checkedfor partial load and performances.

2.7.1.1 Optimization of Engine Configuration . The processes involved in areal Stirling cycle are complex. The interaction between these processesmakes the problem even more complicated. Simulation of a real Stirlingcycle by computer techiiolog;* offers in-depth knowledg'? of the cycle and

MCIUMCM 9 / 1nCHMHooT ^~^ZiNCOll*0ArtO


65/189

Table 2-4

EHIII

I


66/189

can be used to determine quantitatively the effect on the engine perfor-mance of each design parameter. Computer simulation utilizes a set of dif-ferential equations which express the principle of energy conservation, massand momentum based on theories of chermodynamics, dynamics of compressibleflow and heat transfer. These relationships are coo complex for a generalanalytical solution, and hence, they are solved numerically by computer.

By specifying the necessary limitations and operating conditions and the threeengine-alternator configurations selected for the parametric analysis, thecomputer seeks the combined optimum configuration so thac the highest thermalefficiency is achieved. The computerized optimization was performed utilizingthe free-piston Stirling engine optimization computer program developed bySunpower Inc. This highly sophisticated computer program varies all the engineparameters, except those given limitations, until an optimum configuration isobtained as well as any detailed information in any segment of the cycle, ifdesired.

2.7.2 Constraints and Input to the AnalysisFor this analysis, the input and constraints are as follows,

Material Selection and Engineering Propertiesa. Inconel 713 L.C.

A practical Stirling cycle has the same characteristicsas a Camot cycle, the higher the working gas temperaturein the hot space or the lower the temperature in the coldspace the higher the cycle efficiency. The temperatureof the coolant is usually limited by the environmentalconditions. To obtain the highest engine efficiency, itis desirable to operate the engine at the high'ist possibleworking gas temperature. A Stirling engine usually operates.t high pressure and the heat transfer surface is limitedto avoid unnecessary dead volume. In addition, to withstandthe high temperature and pressure, it is also desirable tohave a material v.;h high thermal conductivity. Therefore,the material selection for Stirling design is a major con-sideration. To this end, metallurgists at MTI were con-sulted. After a careful study, Inconel 713 L.C. was^ wetuMCM o / /rtCMHOlUIT ^''iNCOro*no


67/189

selected not only for its high-creep stress out also furits good castabillty and the absence of cobalt, a utrate-gic material.

Therefore, this material's properties were used for theselected engine configurations with the exception, ofcourse, of the ceramic heater head design.

The engineering prc^jerties of Inconel 713 L.C used as inputto the computer program are as follows:

1% Creep Stress for Thermal ConductivityTemperature ("F) 50.000 hrs (psi) Btu/hr ft^/in .

1200 45,000 1661300 36,600 1731400 23,800 1791500 15,400 1991600 11,200 2181800 ?,9002000 1,300

Specific density .28b lb/in. 'Specific heat .14 Btu/lb "F

b. Silicon CarbideCeramic materials were also investigated as possible candidatesfor heater head material. Tney differ considerably from mostmetals in their mechanical properties. Ceramics are consider-ably morf brittle Lhan mosc metalo and consequently failure inceramics occurs less predictably and moie often . castrophic-ally. Silicon carbide was chosen to be used in the engineoptimization program for the selected concept at 1800 and2000"? only. The ceramic heater head engine concept wasincluded in the parametric analysis to indicate the perform-ance potential of such an advanced engine design. However,before ceramic material could be unod as a Stirling engine

2-45rtCMmoofiNcoaKMuno


68/189

compcnent, a considerable amount af fabrication, technology,development and experimental verification of reliabilitywould have to be performed. The properties used in the programwere as follows:

Tensile StrengthCoefficient of ExpansionSpecific HeatDensityThermal Conductivity

General Engine constraints:FrequencyEngine Output

Piston Diameter

Piston .\mplicude

15,000 psi2.7 X lO*'.JO Btj/lb "F.10 lb/in.15 Btu/hr ft "F

Piston Weight

Displaccr DiameterDisplacer Weight

60 Hz5.5 kW (3 units for 3 phases)or 15 kW single unit5" (determined by alternator designwith stator coil inside piston)0.6" ?or kW engine0.69" for 15kW engine(figures determined by alternatordesign)14.3 and 24 lb, respectively (weightsdetermined by alternator design)5" (selected for design simplicity)3.54 lb (determined by geometry)

Hot Space Metai Temperature 1200, 1400, 1600, 1800, and 2000*^Coolant TemperatureNumber of Cooler Tubes

McwuiieurtCMHMMTiNeaaMHATiD

70 and 150''F100 (cooler tube length, diameter andnumber of tubes can be optimized bycomputer program. )ptimized figuresare theoretical not practical formanufacturing. Number of tubes wasmade as an input and computer seekstube diameter and length for maximumthermal *ifficiency)

2-46


69/189

Number of Heater Tuso. 46 same reason as coolerWorking '"as Hydrogen for hlghesc efficiencyRegenerator Porosity .92 based on MTI experience

2.7.3 Analysis ResultsThe results for the 5 kW and 15 kW engine configuration and performance underdifferent temperature combinations are lisr-'d in Table 2-5 through Table 2-7.The Intended optimization of both engines at not space material temperatureof 2000^F was not performed because the thermal efficiency dropped at 1800"?.This is partially due to the drop of the metal creep strength that requires aheavier wall, and increased heat conduction loss from the hot to the cold end.

The highest overall thermal engine efficiency of 43.1 percent was obtained forthe 5 kW engine operating at a 871**C heatec head temperature and a 21''C coolertemperafure. The engine thermal efficiency fur the three engine configurationsat the various heater head and cooler temperatures ai'e presented in Figure 2-17.

The "best" engine configuration (i.e., hijjhsst ef f-'ciency) is run to check en-gine perfor:jance at off-design and partial-load requirements. The partialload results for tl


70/189

TABLE 2-55.5 KW ENGINE AT DIFFERENT OPERATING TEMPERATURES

HEATER TEI1FERAII}RCOOLER TEMPERATURETotal P-V PowerPiscon NET POWERFrequencyCharge PressureHeat InputPiston AmplitudeDlsplacer AaplitudeDisplacer Phase AngleOisplacer Cyl. DiameterDisplacer Rod Oiamscer?lston Frontal AreaRegenerator PorosityRegenerator LengthRegenerator Void VolumeMler (fall TemperatureCooler Tube DiameterCooler Tube LengthCooler TubesAverage Cooler ATHeater Wall TeoperacureHeater Tube DiaaecerHeater Tube LengthSeater TubesAverage Heater ^IEfficiency

5 KW ENGINE1200'F WOO'F

70F ISO'E 70 5^ 150" EL(watt)


71/189

TABLE 2-5 (Cont'd)5.5 KW ENGINE AT DIFFERENT OPERATING TEMPERATITIES

aEAiEa rsMPEaATt-HEccoLsa rESffERArjaElocal ?-V PowerPiscoa NET POWERFrequencyCharge PressureHeac InpucPtscoa Aoplisude3isplacer AaclicudeOispiacer Phase AngleOlsplacer Cyi. Oiamecer31splacer Rod OiamecerPlscon Froncal AreaRegeneracor PorosityRegenerator LengthRegeneracor Void VoluneCooler Wall TataperacureCooler Tube DianecsrCooler Tube LengchCooler tubes.^.vsrage Cooler uTHeater Vail TeaperacureHeacer Tube DiacecarHeater Tuoe LangcUHeater TubesAverage Heater AtEfficiency

:o*'F(watt)(wace)(hz)(bar)(wacc)Cca)Ccat)()CC3)(ca)(ca2)

(ca)(ca3)CO(sa)(ca)(>')C'c;

(=3)(ca)(?)

CO(%)

3 KW EN'GINEIJO'F 70'F LSOO'F 150F


72/189

TABLE 2-615 KW ENGINE AT DIFFERENT OPERATING TEMPERATURES

15 KW ENGINEHEATER TES1FSSA7URECOOLER TEMPERATURETotal P-V PowerPiscoa NET POWERFrequencyCharge PressureHeac InputPiston AoplitudeDlsplacer AoplitudeDlsplacer Phase AngleOisplacer C7I. DlaiaecerDlsplacer Red OlaseterPiston Frontal AreaRegenerator PorosityRegenerator L&ngthRegenerator Void VoluaeCooler 'Mall TeoperatureCooler Tube DiaseterCooler Tube LengthCooler TvibesAverage Cooler uTSeater Vail TeaseratureHeater Tube DiaaeterSeater Tube LengthHeater TubesAverage Heater ATEfficiency


73/189

TABLE 2-6 (Cont'd)15 KW ENGINE AT DIFFERENT OPERATING TEMPERATURES

HEAISR tiMPE3ATURCOOLEX TSJlPSSArjSETotal ?-V Power (wa:Pis con SET PO^^TR (vact)Frequency (hz)Charge Pressure (bar)aeat Input (wacc)Piston Aoplicude (cs)Displacer Amplitude (c3)Jisplacer Phase .Angle (')Olsplacer Cyl. Oiamecer (cs)Oisplacer Rod Olaaecer (ca)Piston Frontal Area (ca^)Regenerator PorositvRegenerator Length (ca)Regenerator Void 7oluste (ca^)Cooler Wall Tenperacure ("OCooler Tube Diaaecer (am)Coder Tube Length (ca)Cooler Tubes (^)Average Cooler II i'C)Heater 'wail Teaperature COHeater Tube Diaaecer (an)Heater Tube Length (ca)Heater Tubes i^).Average Heater AT CC)Efficiency (%)

13 ICU ENGINE1600F

70't 150'F1800F

70"F ISCF)


74/189

TABLE 2-715


75/189

TABLE 2-8PARTIAL LOAD PERFORMANCE OF THE 5.5 KW ENGINE

5 KW ENGINEHSAISR TiMPESATURE loOCF Heacer 70 F CoolerCOOLER TEMPSRATa-RS


76/189

TABLE 2-915 '-W CERAMIC SNGINE 7" PISTON DIAMETER ENGINE

15 KW 7" PISTONHEATER TEIiPERAIURECOOLER TEMFSRAIURElocal P*V PowerPiseoa NET POWERFceqaeseyCharge PressureHeat InputPiston AopllcudeOisplacer AmplitudeOisplacer Phase AngleOisplacer C7I. DiameterOisplacer Rod OiameterPiston Frontal AreaRegenerator PorosityRegenerator LengthRegenerator Void 7oluiseCooler Vail TeameratureCooler Tube DiameterCooler T'lbe LengchCooler TubesAverage Csoler ilTHeater Vail TemperatureHeater Tube DiameterHeater Tube LengthHeater TubesAverage Heacer ATEfficiency

Inconel 713


77/189

50

l^i kW 70^^ COOLER CFWf'K

v."

c0)


78/189

50

40

suuOi

0)30

20

CO

>o 10

20 40 60Percent of Power

80 100

Fig. 2-18 5 kW Engine Partial Power Overall Thermal Efficiency

MICMAMICALTtCIWOKWiNCOiinxuirco 2-56


79/189

2.8 Ranking

2.8.1 MethodRanking of attractive configurations was performed in two stages. Stage1 is a preliminary screening prior to the parametric analysis, used to iden-tify the three most attractive engine concepts and Stage 2 is a screeningafter parametric design and identification of attractive total system concepts.The ranking method used will be a decision analysis technique developed byKepner-Tregoe.

Basically, this method identifies all desirable objectives as either musts orwants. In order to be a viable alternative, a configuration/concept must meetthe must objectives. A preliminary screening of all configuration/conceptsprior to input to the parametric analysis was performed to Insur? all engineconcepts analyzed met the must conditions. These must conditions were:

Capable of 15 kWe power output Capable of engine/alternator efficiency of 40 percent 3 phase - 60 Hertz output

At the completion of the parametric analysis, eich engine configuration wasranked based en the Kepner-Tregoe method. Since all configurations/conceptspassed the preliminary screening, then all met the must conditions. All otherobjetuives are classified as wants and are given relative weighting factorsfrom 1 to 10, with 10 being the most desirable objective. Each configurationthen is reviewed with respect to the ability to comply with objectives rela-tive to all other configurations; the best configuration for each objective isgiven a 10 and all other configurations a relatively low number.

After each configuration was rated relative to each other for each desirableobjective, the relative rating is multiplied by the weighting factor for eachobjective and the total was then summed with the totals for each objective.This results in a numerical rating of all attractive Stirling engine/alternatorsystem configurations with the highest number representing the most attractiveconfiguration/concept

IttCMAMlCAiTICMIKXOa* 257iNCOnvoiuno


80/189

If two or mora configurations were numerically close, Ic was necessary tomake a final choice based on an assessment of adverse consequences of eachconfiguration and a review of how each configuration satisfies the mostweighted objectives.

2.8.2 CriteriaThe established requirements for the Task 1 study are:

power output 15 kW - 3 phasesystem efficiency >, 40 percent

Based on these requirements some Initial engine/alternator configurations(i.e., a free-piston Stirling being used to drive a turbine) were not con-sidered further. The wants (i.e., objectives) that were defined for thisevaluation and their relative importance (weighting factor) are as follows:

Efficiency


81/189

The results of the parametric analyses is presented in Subsection 2.7. Usingthe parametric analyses and engineering judgement, each of the 3 engine c.n-figurations at different operating temperatures were ranked according tooverall system considerations. Hence, while a single 15 kW ceramic heaterhead engine has a higher engine efficiency than any otht-r crsir."? (see Figure2-19), the overall system efficiency is slightly lower than the 3 - 5 kW en-gine/system because a 15 kW system would require a static converter to meetthe required three-phase specification. Obviously if three phase was not arequirement, the 15 kW ceramic heater head configuration would obtain a 10for efficiency and other configurations would be relatively -lower.

Similarly for each want, the system (i.e., collector, selected receiver,engine/alternator/utility grid) is ranked relative to all other systems.The results of this ranking procedure are shown in Table 2-10.

2.8.3 RecommendationBased or the original program guideline of achieving 15 kW-3 phase electricaloutput from a single disk, the 3-5 kW engine/alternator/system operating at871 "C was salected. The single 5 kW engine conceptual layout is shown inFigure 2-19. However prior to initiation of the conceptual design phase ofthis study results of the parametric analysis and configuration definitionwere reviewed with NASA/Lewis. It was determined at that time that the threephase output i-equirement did not need to be obtained with a single dish-enginesystem but could be obtained with a 3-15 kW dish-engine system. Because ofthis change of definition the con-'ept selected for the conceptual design andimplementation assessment was a single 15 kW engine-alternator system.

MtCIUMlCAlrtCMHOiOO*iMOOIKMUnO

2-59


82/189

9HIini2f TABLE 2-10RANKING OF ATTRACTIVE ENGINE/ALTERNATOR SYSTEMS

Io

Musts: 13 KUC Output30 - 60 HZ Outputn Eng/Alt > 35X

Sy tena:Wants-Uel the


83/189

('oolaut Passup.eIIHI

ION

n

Supply PleiiomCas Kuarin}{Drain PJenuin

Plunger M.-iynet Ring

i-splacur(inter ingJ cnuin

k 60Fig. 2->j rive kW Engine-Alternator

(Dimensions in cm)

-i


84/189

3.0 IMPLEMENTATION ASSESSMENTThe purpose of the work completed under this task was to Jetermine the potentialfor development of the 13 kV Solar Engine Concept Design to a production enginestatus and for eventual widespread implementation in solar thermal electricgenerating systems. The main objectives that were addressed are assessmentsof:

Technology Status Producibility Durability Growth Potential

A continual cross-feed of information between this implementation assessmentand the near term conceptual design activity was established in order to pro-duce a positive impact upon the near term design.

3.1 Review of Technology StatusThe objeccivss of the wcjrk ccnplerod anut;- this subtask were to reviev the15 kW Solar Engine cv>ncepc design and:

1) Identify aspects of the design which are critical to its success.2) Identify technologies which are key to the success or failure

of a part.3) Assess whether the design represents current technology, adapta-

tions of current technology, or significant improvements.4) Assess the syt-tem sensitivity 'o changes in component technology.

3.1.JL Frae-Piston Stirling Engines/Linear AlternatorsThe basis for components technology for developing an efficient free-pistonStirling engine power conversion system is well established. Although thethermodsmamic analysis used to predict engine performance is covjplex, theactual operation of the engine/al'jcrnator system is known for its simplicity.Nevertheless, the best overall efficiency is achieved when all of the criticalcomponents are uell matched.


85/189

The potential for increasing the system efficiency beyond current perform-ance levels has been verified with laboratory engine and alternator components.

Inherent design characteristics of gas-bearing supported free-piston Stirlingengines show promise for developing a highly reliable system with a long-lifepotential. Although only a limited amount of operating time has been accum-ulated with the laboratory engines, inspection of the hardware confirms thepotential for developing a power con\/ersion system with long life and highreliability.

3.1.2 Component Technology StatusThe 15 VH preliminary conceptual engine design developed as part of Task 1,shown in Figure 3-1 and outlined by part number in Table 3-1, was reviewedin terms of those operating and design parameters which could sign'.ficantlyimpact engine life, cost, and performance. Parameters such as temperatures,stresses, clearances, complexity, materials, etc. were evaluated for eachengine component. As the result of this evaluation, those coT^ponents whichare considered critical to the engine operation were identified. The keytechnologies upon which the success or failure of each component part relieswere also evaluated. The following definitions were adopted for the purposeof categorizing the status of each of the identified technologies.

State-Of-The-Art . This term refers to the current level of sophisti-cation of a developing technology. A significant amount or ^xperioncehas been accumulated with the technology as configured. The designand operational characteristics of this engine are well known.

.\daptation of Current Technology . The technological concep. has beenproven in similar or analogous hardware systems. Adaptation to thesolar Stirling design wi]l require no basic research in materials ormanufacturing.

Significant Improv^'aent in Technology . The concept Is new or has onlybeen verified i.. a laboratory. Development is requiied in order toidentify a ui design approa '' for adapta-.ion to a Scirling enginesystem.

3-2ToCMi^. OCT


86/189

I

^.** 792030Fig. 3-1 15 kW Preliminary Conceptual Design


87/189

TABLE 3-1IS kW CONCEPTUAL TESIGN PARTS LIST

Icera

123456789

1011

121314151617181920212223

2425262728

Heacer Hepl3cer Spider and PostDisplacer Post-Inner"0" Ring - Displacer Post"0" Ring - Displacer Post

Qty.


88/189

TABLE 3-115 kW CONCEPTUAL DESIGN PARTS LIST (cont'd)

Item


89/189

Table 3-2 Summarizes the critical components and the corresponding assessoientof the technology status.

3.13 Technology Sensitivity AssessmentVariations in the component technologies were assessed in terms of their im-pact on overall system characteristics. The sensitivity of cost, reliability,performance, and maintenance requirements were evaluated for changee in thetechnology status for each critical component. The following outline high-lights that assessment.

I. HEATER HEADA. Sensitive Parameters : Cost, life, reliability, efficiencyB. Independent Parameters ; Design approach

brazed-tube design cast-head design

C. ConsideiaC ions ; The heater head/regenerator housing assembly isone of the most crucial components in the entire engine/alternatorsystem. Because of high engine operating temperatures, high-creepstrength materials must be considered. Wall thickness must be keptto a minimum in order to limit thermal conduction losses, and anadequate heat transfer surface area must be maintained in o^der toeffect the required energy input to the machine. An adequate heattransfer surface with a minimum void volume can readily be achievedvia a brazed-tube heater design. On che other hand, single-casthead designs offer the potential for significantly reducing themachining and assembly time required for the more conventional heater-tube tjrpe designs. In addition, the c

Documents

15kw Free Piston