automatic combustor controller liquid-fuel combustor … · the combustor controller design and...
TRANSCRIPT
AFWAL-TR-82-2120
AUTOMATIC COMBUSTORCONTROLLER DEVELOPMENT ANDLIQUID-FUEL COMBUSTORDESIGN PROGRAM
00( MECHANICAL TECHNOLOGY INCORPORATED
968 Albany-Shaker RoadS) Latham, New York 12110
January 1983
I.
Final Report for Period FEBRUARY 1981 - SEPTEMBER 1982
CLU
I 1Approved for Public Release - Distribution Unlimited
••DTICAeroPropulsion Laboratory JUL I b 1 j•3Air Force Wright Aeronautical LaboratoriesAir Force Systems Command
Wright-Patterson Air Force Base, Ohio
83 07 14 053
NOTICE
When Government drawings, specifications, or oC~.ar data are used for any purposeother than in connection with a definitely related Government procurement !peration,the United States Government thereby incurs no responsibility nor any obligationwhatsoeveri and the fact that the government may have formulated, furnished, ox Inany way supplied the said drawings, specifications, or other data, is not to be re-garded by implication or otherwise as in any manner licensing the holder or anyother person or corporation, or conveying any rights or permission to manufactureuse, or sell any patented invention that may in any way be related thereto.
This report has been reviewed by the Office of Public Affairs (ASD/PA) and isreleasable to the National Technical Information Service (NTIS). At NTIS, it willbe available to the general public, including fc eign nations.
This technical report has been reviewed and is approved for publication.
A
VALERIE /VAN GRIETHUYSEW PAUL R. BERTHEAUDProject Engineer Chief, Energy Conversion BranchEnergy Conversion Branch Aerospace Power Division
hf Aerospace Power DivisionAero Propulsion Laboratory
ii
"If your address has changed, if you wish to be removed from our mailing list, orif the addressee is no longer employed by your organization please notifyAFWAL/POOCN-PAPB, O 45433 to help us maintain a current mailing list".
Copies of this report should not be returned unless return is required by securityconsiderations, contractual obligations, or notice on a specific document.
UNCLASSIFIEDSECURITY CLASSIFICATION OF THIS PAGE (*Woien DV.44tared•,
REPAORT DOCUMENTA.TIO READ INSTRUCTIONSBEFORE COMPLETING FORM
T. REPORT NUMIOER 1 A N RECIPIENT'S CATALOG NUMBER
AFWAL-TR-82-2120 o eA4. TITLE (and Subtitle) S. TYPE OF REPORT A PERIOD COVERED
Automatic Combustor Controller Final Report for PeriodDevelopment and Liquid-Fuel Combustor Feb 81 - Sep 82Design Program 6. PERFORMING O1G. REPORT NUMBER
7. AUTHOR(s) S. CONTRACT OR GRANT NUMBER(s)
J.R. Killough T.M. MoynihanR.A. Lacey R.A. Farrell F33615-81-C-2007
9. PERFORMING ORGANIZATION NAME AND ADDRESS 10, PROGRAM ELEMENT, PROJECT, TASKAREA & WORK UNIT NUMBERSMechanical Technology Incorporated Program Element 62203F
968 Albany-Shaker Road Project 3145, Task 314524Latham, New York 12110 Work Unit 31452433
I1. CONTROLLING OFFICE NAME AND ADDRESS 12. REPORT DATE
Aero Propulsion Laboratory (AFWAL/POO) January 1983Air Force Wright Aeronautical Laboratories 13. NUMBER OF PAGES
Wright-Patterson Air Force Base, OH 45433 18514. MONITORING AGENCY NAME A ADORESS(ifdifferent from Controlling Office) IS. SECURITY CLASS. (of this report)
Unclassified 1I5a. DECLASSIFICATION/DOWNGRADING
SCHEDULE
16. OISTIIIBUTION STATEMENT (of this Report)
Approved for public release: Distribution unlimited
17. DISTRIBUTION STATEMENT (of the abetract entered In Block 20, i( different from Report)
18. SUPPLEMENTARY NOTES
"The appendix to this report has been intentionally omitted because itcontains a computer software program. Requests for this document must bereferred to AFWAL/POOC, Wright-Patterson AFB, OH 45433. Non-DoD request
t•tIjclude the statement of terms and conditions contained in Atch 21 to AFR19. KEY WOROS (Continue on reverse aide if necessary and Identify by block nL'tber)
Free-Piston Stirling Engine Logistic FuelsCombustor Controller Mutlifuel CapabilityControl System Monlithic Heater HeadCombustorLiquid-Fuel Combustor
70. ABST T (Continue on reverse side it neces•a•y and identify by block number)
"This technical report, prepared for the Aerospace Power Division, AeroPropuliosn Laboratory, Wright-Patterson AFB, states design and test resultsof a combustor controller integrated with an existing 1kW free-pistonStirling engine. This report document also states the design of a liquid-fuel external heat system capable of operating with a 3kW free-pistonStirling engine.
DD I JAN 1473 EDI'ION OF I NOV 6S 1S ODSOLETE UnclassifiedSECURITY CLASSIFICATION OF THIS PAGE (When Data Entered)
- _ _s
UNCLASSIFIEDSECURITY CLASSIFICATION OF THIS PAGI(When Date Entered)
The combustor controller design and development is directed at a con-trolled system that maintains the natural gas fired Stirling engine heaterhead temperature. In response to either load and/or stroke change of theStirling engine, the basic controlled strategy, control system functions andsystem hardware is documented. The results and conclusion of steady stateand transient testing of the controller with a free-piston Stirling engineare presented.
A liquid-fire external heat system includes evaluation of a mol'ilithicfree-piston Stirling engine GPU-3 heater head diesel combustor nozzle anddesign of a liquid-fueled external heat system that can integrate "ith a 3kWfree-piston Stiling engine.
Aooesston ForNTIS GRA&IDTIC TAB
Justif i cnt t
By_ . ...
vni :'.-"i i y 'odes
1', ,
UNCLASSIFIEDj SECURITY CLASSIFICATION OF 'Uwe PAGE(Wben Data Entered)
PREFACE
This document contains the results of work performed for Wright-Patterson Air
Force Base under Contract No. F336'S-81-C-2007. The work effort was performed
over two separate time periods. li=m February 1 through November 30, 1981, an
automatic combustor controller was designed, fabricated, developed, and tested.
The design of a potential multiliquid-fuel. combustor that would integrate with a
free-piston Stirling engine (FPSE) was performed from April 1 through September
30, 1982. KThe major authors of this report, and their contributions are presented below:
Joseph Killough - definition/design of the automatic combustor
contruller;
Robert Lacy - fabrication, software, and development of the
automatic combustor controller;
Thomas Moynihan - integration/test of the automatic combustor
controller; baseline testing of an existing
FPSE; and,
Roger Farrell - design o. the multiliquid-fuel combustor.
As expected, the development of an automatic combustor controller and a
liquid-fuel combustor for an FPSE provided the first step towards an integrated,
militarized FPSE to be used for stand-alone, tactical field power.
Development efforts of this promising energy-conversion system are continuing
under separate contracts.
L lii
TABLE OF CONTENTS
Section Page
INTRODUCTION ...................... ................................ 1-1A. Background .............................................. ....... 1-1B . Objective ..................................................... 1-1C . Scop e ......................................................... 1-2D . Summ ary ....................................................... 1-4
1. Combustor Controller ...................................... 1-5.-52. Liquid-Fuel Combustor Design .............................. 1-10
II COMBUSTOR CONTROL SYSTEM DEVELOPMENT .............................. 2-iA, Characterization of Existing Manual Control System ............ 2-1
1. Performance Characterization .............................. 2-32. Transient Characteristics ....... .......................... 2-33. Fuel/Air Loop Characterization ............................ 2-25
B. Design, Fabricate, and Develop Automatic Control System ....... 2-301. Basic Control Strategy .................................... 2-302. Control System Functions .................................. 2-303. Control System Specifications ............................. 2-324. System Software ........................................... 2-40
III EVALUATION OF FPSE INTEGRATED CONTROL SYSTEM ...................... 3-1A. Engine Load Control ........................................... 3-1B. Start/Stop Sequence ........................................... 3-2
IV SYSTEM INTEGRATION ................................................. 4-1
V TEST AND EVALUATION ............................................... 5-iA. System Test of Combustor Controller ........................... 5-1
1. Steady-State Tests ........................................ 5-i2. Transient Tests ............................................ 5-13. Conclusions ............................................... 5-11
VI LIQUID-FUEL COMBUSTOR/EXTERNAL HEAT SYSTEM EVALUATION AND DESIGN.. 6-iA. Evaluation of Monolithic Heater Head .......................... 6-1
1. Performance of TDE Heater Heal ............................ 6-i2. Suitability of Monolithic Hea'.er Head for Air Force
Logistic Fuels ............................................ 6-4B. Emissions of the TDE External Heat System...................... 6-6C. GPU-3 Diesel-Combustor Test and Baseline Evaluatioi ........... 6-13D. Liquid-Fuel Combustor Cup and Nozzle Design ................... 6-27E. Conclusions of Liquid-Fuel ExteLnal Heat System Design ........ 6-32
VII CONCLUSIONS/RECOMMENDATIONS ......... ............................. 7-i
VIII REFERENCES ........................................................... 8-1
V-4 EHEDL O BL N-NOT nL=
LIST OF FIGURES
Number Page
1-1 Technology Demonstrator Engine Layout ............................... .1 31-2 Automatic Combustor Controller ................................... i-s1-3 Interior of Automatic Combustor Controller .......................... 1-71-4 Input/Output Signals to Automatic Combustor Controller .............. 1-81-5 Automatic Combustor Controller Installed in MTI TDE/FPSE Test
Facility .......................................................... 1 +91-6 Liquid-Fueled Combustor and Fuel Nozzle ............................. 1-112-1 FPSE Heater Head Temperature Control ................................ 2-22-2 Total Cycle Power vs Piston Stroke and Heater Control Temperature... 2-62-3 Total Cycle Efficiency vs Piston Stroke and Heater Control
Temparature ....................................................... 2-72-4 Displacer Stroke Ratio vs Piston Stroke and Heater Control
Temperature ....................................................... 2 -82-5 Displacer Phase Angle vs Piston Stroke and ;:evtex Control
Temperature ............................................. ......... 2-92-6 Firing Rate v- Piston Stnoke and Heater Control f&,iyK'rature ......... 2-102-7 Heat Input to Engine vs Piston Stroke and Heater tkontrol Temperature 2-112-8 Heater Head Base Temperature Profile ................................ 2-122-9 Air/Fuel Ratio vs Piston Stroke ............................ ........ 2-112-10 Mean Heater Temperature vs Time (Fixed Firing Rate) ................. 2-142-11 Control T/C Temperature vs Time ..................................... 2-152-12 Total Cycle Power vs Time ........................................... 2-162-13 Total Cycle Efficiency vs Time ...................................... 2-172-14 Piston Stroke vs Time ........................................... 2-182-15 Average Heater Head Temperature vs . ime ............................. 2-192-16 Control T/C Tomperature vs Time ........................... .......... 2-202-17 Total Cycle Puwer vs Time ........................................... 2-212-18 Total Cycie Efficiency vs Time ...................................... 2-222-19 Burner Firing Rate vs Time ............................. 2-232-20 Air-To-Fuel Volume Flow Ratio vs Time ............................... 2-242-21 Airflow vs Valve Input of 6 V 0.1-Hz Triangle 0.25 V/cm
Vertical and Horizontal.. .. ....................................... 2-262-22 Fuel Flow vs Valve Input of 6 V 0.1-Hz Triangle 0.25 V/cm
Vertical and Horizontal ...... _..................................... 2-272-23 Airflow/Fuel Flow vs Time Valve Input - 0.2-Hz Square Wave .......... 2-202-24 Airflow vs Time Valve Input ............................ ............ 2-292-25 WPAFB Air/Fuel Controller Control Loop Structure .................... 2-312-26 WPAFB Air/Fuel Controller System Block Diagram ...................... 2-333-1 Engine Start-Up ..................................................... 3-33-2 Combustor Start Sequence ............................................ 3-43-3 Engine Stop Sequence ................................................ 3-65-1 System Power vs Piston Stroke ....................................... 5-35-2 Total Efficiency vs Piston Stroke ................................... 5-45-3 Heat Load vs Piston Stroke ................. ....................... 5-55-4 System Firing Rate vs Piston Stroke ........ ...................... 5-65-5 Combustor System Efficiency vs Piston Stro. ..................... 5-75-6 Controller Mean Temperature and Piston Stroke vs Time............... 5-95-7 Various Parameters vs Time .......................................... 5-105-8 Various Parameters vs Time ......................................... 5-125-9 Various Parameters vs Time ......................................... 5-135-10 Various Parameters vs Time .......................................... 5-146-la TDE Gas-Fired Combustion System ................................... 6-26-1b TDE Finned Heater Head.............................................. 6-26-2 Heater Head Temperature Axial Distribution...........................6-36-3 Average Heater Head Temperatire vs Piston Stroke.................... 6-56-4 TDE Emissions - Lambda vs 3aseowi Emissions ......................... 6-76-5 TDE Emissions - Lambda vs Emissions Yndex ........................... 6-8
Vi
II
LIST OF FIGURES (Cont'd)
Number Page
6-6 Exhaust Temp/Preheat Temp/Burner 9 vs Air/Fuel Ratio ............... 6-96-7 Heater Head Load/Max Fin Temp/Head Exit Gas Temp vs Air/Fuel Ratio. 6-106-8 Area Weighted Mean Head Temperature and Firing Rate vs Air/Fuel
R atio ............................................................ 6 -116-9 GMR Two-Piece Burner Assembly ...................................... 6-146-10 GMR Fuel Nczzle and Igniter Unit ................................... 6-146-11 Free-Piston Combustion Rig ......................................... 6-156-12 GPU-3 Combustor Evaluation (Diesel Fuel) ........................... 6-196-13 GPU-3 Combustor Evaluation (Diesel Fuel) ........................... 6-206-3.4 GPU-3 Combustor Evaluation (Combustor Cup Thmperature Between
Swirl Slots) (Diesel Fuel) ....................................... 6-216-15 GPU-3 Combustor Evaluation (Diesel Fuel) ........................... 6-226-16 GPU-3 Combustor Evaluation (Diesel Fuel) ........................... 6-236-17 GPU-3 Combustor Evaluation (Diesel Fuel) ........................... 6-246-18 GPU-3 Combustor Evaluation Combustion Gas Temp vs Axial Distance
(T.P. #6) (Diesel Fuel; 15 kW) ................................... 6-256-19 GPU-3 Combustor Evaluation Combustor Pressure Drop (Diesel Fuel)... 6-266-20 Liquid-Fueled Combustor and Fuel Nozzle ............................ 6-286-21 TDE Combustor Cup .................................................. 6-296-22 Sono-Tek Fuel Nozzle ............................................... 6-316-23 Proposed EM Diesel Fuel Nozzle and Combustor Cup ................... 6-33
LIST OF TABLES
Number Page
2-1 Performance Map Test Matrix ........................................ 2-4z-2 Summary of Transient Test Points ................................... 2-52-3 Combustor Controller Specifications ................................ 2-345-1 Steady-State Test Point Matrix ..................................... 5-25-2 Transient Test Point Matrix ........................................ 5-86-1 GMR GPU-3 Engine Dynamometer Performance ........................... 6-176-2 GPU-3 Combustcr Evaluation Test Results ............................ 6-18
Vii
I. INTRODUCTION
This is a final report of the results of an eight-month contract awarded to
Mechanical Technology Incorporated (MTI) by the Wright Patterson Air Force Base
(WPAFB) Aero Propulsion Lab. The objective of this program was to develop and
test components to adapt a free-piston Stirling engine (FPSE) for Air Force use
at unattended remote and mobile sites. Emphasis of the work performed under
this contract was centered on the development of an automatic combustor control
system, and the design of a liquid-fueled combustor system that could eventually
lead to unattended site operation.
A. OBJECTIVE
The major objective of this Automatic Combustor Controller Development and
Liquid-Fuel Combustor Design Program was the development of an automatic
combustor control system (a significant step towards the integrated
engine-alternator control system) that would allow for unattended operation of
an FPSE at Air Force remote and mobile sites. A second objective was the analyt-
ical evaluation and design of a multifuel combustor/heater head sucl that one
engine system could use a variety of military logistic fuels.
B. BACKGROUND
The FPSE s an attractive power-conversion system because of its potential for
long life, high reliability, and multifuel capability. A l-kW FPSE/linear
alternator, designed, fabricated, and i -sted by MTI under a separate program
sponsored by the Department of Energy (DOE), has been modified to incorporate a
natural-gas-fired combustor system and a monolithic-finned (tubeless) heater
head*. The power conversion system has only two moving parts, whose motion is
determined by their respective mass and gas springs. There is no physical
connection between the dynamic components; the displacer and power pistons are
supported by hydrostatic gas bearings with noncontacting clearance seals. The
natural-gas combustor system has a recuperator to recover exhaust gas thermal
energy, and a turbulent-mixiitg combustor in which natural gas and air are mixed
*Technology Demonstrator Engine (TDE)
1-1
and burned. The Inconel 617 heater head has both internal and external fins to
augment transfer of thermal energy from the hot combustor gases to the internal
helium working fluid. A layout of the TDE power conversion system is prosented
in Figure 1-1. This engine system has been designed to provide 1.4-kW (engine
P-V power) and 1.0-kW electric output at 40-Bar charge pressure. As discussed
in detail in References 1 and 2, the system has previously achieved:
"* demonstration of the l-kWe steady-state design power output;
"* validation of analytical code predictions with test results; and,
"* accumulation of 300 hours of operational test deta.
The TDE is presently being used to demonstrate free-piston Stirling technology
and to develop the FPSE so that its potential for a wide variety of applications Ican be realized. Demonstration of an automatic combustor control system, and a
previous evaluation and design of a multifuel combustor heat system, enhances an
earlier demonstration of a military FPSE. The automatic control system will be
integrated with the TDE, and the liquid-fuel combustor design will be integrated
with the FPSE* presently being developed by MTI.
C. SCOPE
The program was divided into five discrete elements to meet the overall program
objectives. Detailed technical discussions of each major task are presented in
Sections II through VI, respectively, of this report.
"* Task I - Combustor Controller Characterization and Development -
The purpose of this task is to specify, design, fabricate, and
component-test a combustor controller that will maintain FPSE heat-
er head temperatures with respect to varying airflow/fuel flow
and/or Xoad changes.
"* Task 2 - Evaluation of Integrated FPSE Control - The combustor
controller is one major element of an integrated system control that
will provide unattended FPSE operation; the other is an FPSE power
*has similar features to the TDE, but has greater power capacity (3.0 kWe)
1-2
AirFuel Nozzle
Inlet
= • ~ N----mExhaust
PreheaterCombustion
Expansion • ChamberSpace ----- Burner Liner
Heater -Heater Head(Internal)
.____ __•..----DisplacerRegenerator
p
Cooler
44
Displacer
Rod CompressionDisplacer Spaces
Gas SpringCompression SpaceConnector Duct
Bearing__Plenum S.-• Power Piston
P e.. -- Alternator PlungerPressure•Vessel
"Alternator Stator
BearingPlenum•.m
30- 'Bounce Space
PistonGas Spring
Figure 1-1 Technology Demonstrator Engine Layout
1-3
control system that will maintain engine operation during load
transient. A cursory evaluation of FPSE power control is performed.
"* Task 3 - System Integration - The combustor controller developed in
Task 1 is integrated and checked out as part of the FPSE control and
test facility operation.
ti"* Task 4 - Evaluation - The system is tested to ejaluate combustcr
controller response to load changes.
"* Task S - Liquid-Fuel Combustion Design - The potential of the FPSE
will be analytically evaluated for multifuel operation. Based on
the current netural-gas combustor technology, a design of a
liquid-fuel combustor will be developed.
D. SUMMARY
The major directed thrust of this program, the development of an automatic
combustor control system for an existing iaboratory FPSE, was completed with
great success. The remainder of the program studied other potential problems
(including the design of a liquid ruel system) whilb using FPSE's at Air Force
unattended sites. The major conclusions are:
• development of an engine power control is necessary to provide
load-following capability for unattended operation;
o engine power control can and will interface with the WPAFR combustor
controller;
* a control system can be developed to provide for ,inattended
operation;
• a monolithic heater head such as used on the TDE can burn military
logistic fuels;
* emissions of the present natural-gas-fired system ara very low;
1-4
* development of a liquid-fuel system is required;
* design of a liquid-fuel combustor with multifuel capability is
possible; and,
an engine/alternator system can be developed to meet the require-
ments of the Air Force for unattended operation.
1. Combustor Controller
The combustor controller developed in this program performed as expected by:
"• automatically mair.aining hrater head temperature with respect to
air/fuel rates and/cr engine load changes;
"* providing start-up interlocking with the TDE facility;
"* providing close-loop control of air/fuel ratio and temperature mean
max with operator set capability;
• displaying key operating parameters;
* providing automatic shutdown; and,
* providing expansion capability to interiace with power control
system.
The demonstration range of performance parameters includes:
* firing rate 3-10 kW
9 temperature 200-5000C (error of only +10 0 C)
9 system output power 200 watts to 1 kW
* air/fuel ratio 15:1 to 35:1
The combustor controller hardware is shown in Figures 1-2 to 1-4, and installed
in the TDE facility in Figure 1-5.
1-5
_-rn-A
II
0
4-i
0
U
0
gU
C-,
C.',
'-4
-'--4
4
1-6
�
UJ
0
1-7
"4-tMr4-j
CU~A 41,.. A~~~t ~ - 15
A tg- ~ 4w , *4 ý4I
1-84
- .. -. .....-.
4I A
V4J
Nix A;
i-A 0
~ i ~ ~ nor - ---
2. Liquid-Fuel Combustor Design
The design of a liquid-,fuel external heat system (shown in Figure 1-6), based on
existing natural-gas combustor/external heat system technology, has the capa-
bility for multifuel use. The design of this combustor will be fabricated,
tested, and developed under separate funding. This combustor design integrates
with MTI's existing 3-kWe FPSE, designated as the Engineering Model (EM).
I--I0 iI"
F'ael Nozzle
Combustor Cup
Ceramic Insulation .
Ceramic Flame Shield
MonolithicHeater Head
Ceramic Fin Stuffer in).
'. ;-ii -
-7 T
I'
Figure 1-6 Liquid-Fueled Combustor and Fuel Nozzle
I-ii 822605
II COMBUSTOR CONTROL SYSTEM DEVELOPMENT
A. Characterization of Existing tanual Control Systems
The manual control system utilized on the l-kW TDE incorporated two
pneumatic-driven valves for gas and airflow, respectively. An operator monitors
the heater head temperatures while manually adjusting air and gas (or both
flows) to maintain a given temperature with respect to changing load require-
ments. A schematic of the present combustor control system is shown in Figure
2-1, along with the planned interface of the automatic system.
A desirable first step to moving the FPSE from the laboratory to an unattended
operation site is automation of the combustor control system. This control
system will utilize a microprocessor control that will provide bot-. autoruatic
and manual functions, with all basic control signals routed through the system,
allowing for easy future expansion of the system, and closed loop control of the
air/fuel ratio in the manual mode. In the automatic mode, heater head temper-
ature will be used to control the firing rate of the combustor. Automatic
shutdown will be provided in the event of relevant F;qsten problems. The basic
signals and functions in the existing facility's mi-,-al control system will be
contained in the new electronic combustion contrQ1 7-y-tem. The existing manual
control will be electrically interchangeable with the new control system, allow-
ing it to be used as backup. In ocder to define the requirements for the
automatic system, determination of the characteristics of the present manual
combustor control system was necessary. These include:
"* system firing rate range;
"• heater head temperature distribution;
"• engine power versus heater head temperature at various strokes;
"• airflow/,fuel flow:
- flow versus valve current (DC),
- step responoo,
- frequency response,
- hysteresis;
" temperature versus air/fuel ratio; and,
"* engine/combustor response to step-load change.
2-1
-w - - - - -- --
Ii
V.
$J3/�L � (ar) -OT H0I-I
[�144
II KIIi
QJ
41 0 p.1909
� 0��
r3-1
-�
(NI
I..V. 4-.-4 0
4444 � '-:44 4-. 99
z �.
41. 4�
� 4� 9 0.M 0I..U �
2-2
The objectives of these tests are both steady-state and trarIsiqnt performance
characterizations, with particular attentiun to heater head performance and
transient heater/engine responses. Testing was performed with the TDE and test
facility. The TDE configuration during combustor characterization testing was:
* 29.93-mm chrome-oxide rod versus hardened steel bearing;
0 gas-fired finned heater head;
• TDE/alternator and piston spring assembly;
0 screen regenerator; and,
* gas-fired combustor and preheater with monolithic heater head.
1. Performance Characterization
The TDE was run at the test zonditions given in Table 2-1 to obtain the steady-
state performance characteristics that correspond to the transient characteri-
zation points given in Table 2-2. Total cycle power and efficiency trends
(shown in Figures 2-2 and 2-3) behaved as expected. At low strokes, the load
stabilization control did not have sufficient gain to maintain stable operation,limiting low stroke operation. At high strokes, heater head characteristics
limited the heat input, limiting high-stroke operation.
The dynamic response of the displacer (shown in Figures 2-4 and 2-5) indicates
normal engine operation. The firing rate and heat input to the head are showq in
Figures 2-6 and 2-7. The heater temperature distribution (Figure 2-8) shows the
heater wall temperature dropping along the length of the head. This drop, an
effect of the external fin characteristics, results in less heater head capabil-
ity than if the temperature were uniform. These steady-state performance tests
were necessary to define, set, and check steady-state control parameters.
2. Transient Characteristics
The transient test matrix given in Table 2-2 was run to evaluate engine charac-
teristics and identify response parameters for the combustor/heater system
controller. The measured responses indicate that the system time constant is
-4-5 minutes (significantly longer than the design response of the controller,
which updates readings once per second). Transient response of the TDE system
is presented in Figures 2-9 through 2-20. The transient system response did not
2-3
TABLE 2-1
PERFORMANCE MAP TEST MATRIX
Heater Temperature
Piston Stroke (T = 0C)
400 500 550
1.4 1* 6* 11*
1.6 2 7* 12*1.8 3 8 13* I2.0 4 9 14
2.2 1 _5 1 i10 1 15
*Engine would not operate at given conditions.
I2-4
TABLE 2-2
SUMMARY OF TRANSIENT TEST POINTS
Test Point Initial Conditions Control Transient
Heater Piston Firing Piston
Temperature Stroke Rate (kW) Stroke
2 400 2.2 +1.0 Const.
3 500 2ý2 +1,0 Const.
4 500 2.2 -1.0 Const.
5 400 2.2 - .75 Const,
6* - -
7*--
8 372 2.2 Const. -. 2
9 425 2.0 Const. +.2
10* --
11 500 2.2 Const. -. 2
12 500 2.0 Const. +.2
-I
*Engine would not operate due to low temperature.
2i
2-5
1800
P total1700-
1600
1500 +
Heater1400 Control
Temperature1300 o
Key: a. 4060"C bb. 500°C
1.200 c. 550 0 C
S1100
S1000
S800 F ~* a700
600
500
400
300
200
100
.0 .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.0 4.2 2.4 2.6
Piston Stroke (cm)
Figure 2-2 Total Cycle Power Versus Piston Strokeand Heater Control Temperature
2-6
- -.
40,0
HeaterControl
35.0 Ptotal Temperature
Key: a. 400 0 Cb. 500 0 Cc. 550°C
30.0
• d ÷C
2 5 .0 -.
20.0
A-I
15.0
10.0,
5.0*
i0 P .
.0 .2 .4 .6 .8 1.0.2 2 1.4 1.6 1.8 2.0 2.2 2.4 2.6
Piston Stroke (cm)
Figure 2--3 Total Cycle Efficiency Versus Piston Strokeand Heater Control Temperature
2-7"~1~
2.0
1.8
1.6
0Ido H e a t e rControl
1.4 Temperature
00
Key: a. 400 0C50%b. 500C"-- 550oc
o 1.2
S1.0
.8
.6
.4
KI
.2
•.0 I I'
.0 .2 .4 .6 .8 1.0 1.2 1.4 1 6 1.8 2.0 2.2 2.4 2.6
Piston Stroke (cm)
Figure 2-4 Displacer Stroke Ratio Versus Piston Strokeand '4eater Control Temperature
2-8
80
70
Heater I S!ControlTempera ure a a
Key: a. 400°C
b. 500°Cc. 550PC
" 50
3• 0
20
10
.0.0 .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.b
Piston Stroke (cm)
Figure 2-5 Displacer Phase Angle Versus Piston Strokeand Heater Control Temperature
2-9
10.0
9.0
HeaterControlTemeratureKey: a. 400uC
b. 500P Cc. 550°C
b7.0 b
S6.0 b
5.0
be4.0 a
aa
3.0
2.0
1.0 HIi
.0.0 .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6
Piston Stroke (cm)
Figure 2-6 Firing Rate Versus Piston Stroke
and Heater Control Temperature
2-10
I.I01.10
1.9 HeaterControlTemperature
1.8 Key: a. 400 0 Cb. 500 0C. 550°C
1.7
:1.6
0
A !
r 1.5
1.3
1.2
1.0
.01.•0 .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6
Piston Stroke (cm)
Figure 2-7 Heat Input to Engine Versus Piston Stroke Hand Heater Control Temperature
2-11 !
Ad9
-44-
-4
00
m~
ii29ada ase
2-12
40
35-
Heater30 Control
Temperature
Key: a. 400'0Cb. 5000Cc. 5500C
25
0 "
0) 20
"-44
15-
10C
5-
.0i
0 .2 .4 .6 .8 1,0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 I!
Piston Stroke (cm)
Figure 2-9 Air/Fael Ratio Versus Piston Stroke
2-13
8O0
700 Key: A - 400 C Initial Control T/C Stroke 2.2-2.0B - 4250 C Initial Control TIC Stroke 2.0-2.2C - 500°C Initial Control T/C Stroke 2.2-2.0D - 500°C Initial Control T/C Stroke 2.0-2.2
0 600
S500.
-- &'• ' -. •;- -"-- - ---- - -
300 B
200
100
.0 10 20 30 40 50 60 70 80
Time (min)
Figure 2-10 Mean Heater Head Temperature Varsus Time (Fixed Firing Rate) I2-14k_, i]
800
Key: A - 400°C Initial Control T/C Stroke 2.2-2.0B - 425 C Initial Control T/C Stroke 2.0-2.2C - 5000 C Initial Control T/C Stroke 2.2-2.0
700 D - 500 C Initial Control T/C Stroke 2.0-2.2
600r
1 500*
. 400
E-0
300
200
100
.0 10 20 30 40 50 60 70 L0
Time (min.)
Figure 2-11 Control T/C Temperature Versus Time
2-15
Sh . . .. . . .. . . . ...
180
Key: A - 400°C Initial Control T/C Stroke 2 2-2.0
160Ky B - 4250 C Initial Control T/C Stroke 2.V-2.2
C - 50 0°C Initial Control T/C Stroke 2.2-2.0- 500°C Initial Control T/C Stroke 2.0-2.2
1400, C
,-.120% :•'
10O A N'160 .4L • ,.- 0 .-- ',
4i00(
oI
0
I I
.0 10 20 30 40 50 60 70 80
Time (min.)
Figure 2-12 Total Cycle Power Versus Time
2-16
40
3C
5 Key: A- 400 0 C Initial Control T/C Stroke 2.2-2.0B - 4250 C Initial Control T/C Stroke 2.0-2.2C - 500"C Initial Control T/C Stroke 2.2-2.0 2D - 5000 C Initial Control TIC Stroke 2.0-2.2
30 LIN
25
20i Ai, "• F-÷
15
10
.0 10 20 30 40 50 60 70 80
Time (min.)
Figure 2-13 Total Cycle Efficiency VWrsus Time
2-17
2.4
CA
Key: A - 400 C Initial Control TIC Stroke 2.2-2.0 U1.6 B - 425 C Initial Control T/C Stroke 2.0-2.2
- 500°C Initial Control T/C Stroke 2.2-2.0SD -500°C initial Control T/C Stroke 2.0-2.2
01.4
o 1. 2
1.0 i
.6
.4
.2
0 _ I
.0 10 20 30 40 50 60 70 80 1Time (min.)
Figure 2-14 Piston Stroke Versus Time
2-18
800
700 Piston Stroke - 2.2 cm
Key: a - 400*C Initial Conditions + 1.0 kW Qin
b - 5UOC Initial Conditions + 1.0 kW Q500C Initial Conditions - 1.0 kWin
600 d - 400*C Initial Conditions - .75 kW Qin
500,
cz II I
] 200
÷b
"100Ix
S 300, d -=•"
200'!
0ý-.010 20 30 40 50 60 70 80
TIME (min)
Figuire 2-15 Average Heater Head Temperature Versus Time
2-19
?
Piston Stroke - 2.2 cm
Key: a - 400*C Initial Conditions + 1.0 kW Qin
b - 5000 C Initial Conditions + 1.0 kW Qin
700 c - 500°C Initial Conditions - 1.0 kW Qind - 400*C Initial Conditions - .75 kW Qin
600
500.
400
-d
300
200
.100
01•0 10 2O 30 40 so 60 70 80 j
TIHE (min)
Figure 2-16 Control T/C Temperature Versus Time
2-20
1800
1600 b~ -
Piston Stroke 2.2 cm
Key: a - 400*C Initial Conditions + 1.0 kW Q1400 b - 500*C Initial Conditions + 1.0 kW Q
c - 500'\' Initial Conditions - 1.0 kW Q n
d - 400%C Initial Conditions - .75 kW Q
12000
14~
C..)
-~800
600
•~ dc
4800 . . /,
200 i
0.0 10 20 30 40 50 60 10 80
TIME (min)
Figure 2-17 Total Cycle Power Versus Time
2-21
4O
35Piston Stroke - 2.2 im
Key: a - 4000C Initial Conditions + 1.0 kWb - 500*C Initial Conditions + 1.0 kW Q
30 c - 500*C Initial Conditions - 1.0 kW Qin
d - 400*C Initial Conditions - .75 kW Qin
ib
Ca. /
20 !jNc
22(1
120
.0 12030 45060 70 80
Figure 2-18 Total Cycle Efficiency Versus Time
2-22
4k__
10 Piston Stroke = 2." cm
Key: a = 400*C Initial Conditions + 1.0 kW Qin
b - 500*C Initial Conditions + 1.0 kW Qin
c -500*C Initial Conditions - 1.0 kW Qin
d - 400C Initial Conditions - .75 kW Q
-4 4
6pp
41
5ý
4
3.r1
21
.0 10 20 30 40 50 60 70 80 1TIME (min),i
Figure 2-19 Burner Firing Rate Versus Time
2-23
40
35 T/C Ratiod
ddd 9dddAddd
i • ~30•"
Piston Stroke 2.2 cm 1
0 25 Key: a - 400*C Initial Conditions + 1.0 kW Q
b - 500*C Initial Conditions + 1.0 kW Qin
c - 500eC Initial Conditions - 1.0 kW Qo0" d - 400*C Initial Conditions - .75 kW Qin
ini
~20
Ii
15
10
5 P
0.0 10 20 30 40 50 60 70 80
TWE (min)
Figure 2-20 Air-To-Fuel Volume Flow Ratio Versus Time
2-24
require modification of the manual control hardware; however, testing showed
heater head temperature to be the control parameter that could be used as the
combustor control. Air/fuel flow could be adjusted automatically to maintain
heater head temperature within a necessary control limit. Heater head temper-
ature response was much longer than that of the airflow and/or fuel flow calves.
3. Fuel/Air Loop Characterization
Fuel/air loop testing was necessary to determine the response rates of the
air/fuel valves, valve setting with respect to input voltage, and hysteresis in
the valves themselves. Air/fuel loop testing ia accomplished with:
"* engine/combustor fully assembled;
"* system at room temperature;
"* flow restrictor on combustor outlet yieldirg 8-1/2" H20 pressure
drop to simulate hot combustor;
• nitrogen flowed in fuel loop (upstream of 40" H2 0); and,
"• function generator connected to controller to modulate valves
(full-scale valve motion caused by 0 to 10 V signal ontput)*.
Typical air/fuel valve transient response is shown in Figures 2-21 through 2-24.Based on transient testing of the air and fuel loops, flow versus valve input
voltage characteristics are:
Qair/V = 1.62/(0 + S/4.1) ± 1.2 cfm/volt for air; and,
Qfuel /V = 0.027/(l + S/3.77)(I + S/5.03) ± 0.013 cfm/volt for fuel;
where: Qair = airflow (cfm);Qfuel = fuel flow (caf);
V = voltage; and,
S valve position.
*controller's voltage-to-current converters were used.
2-25
is.~
it, it
3, :41; - I~ .1-.F.I
-
- ~ ~ ~ ~ ~ ~ ~ ~ -I ;1 !IAlId .,
Vr p:I IT
11: 1 'T - . . .j1*11 A W . 0!
!U~~ 26t i
i T,
,Ii7 h T-.tI1 71 W
U]': lis 4
-:n fr wiIlF.E.VE i- 7 11:'..1 1w Fv 4:
"UL n "Hý41 i*711 4ý +111"
9T Li!Aw~i: ;0
i~~~~' 1!:iilv iiW
P-14- i,4i.~~~~ 71 1M cI~ v`-4
J~c
rr t
+ 0
M~ii F T 1
I~~s M.fwif T",
fl2-27
U. I'f
T R . . 4. ;:1:.
11J.- ... ... .III .r ~ . II 117711 . ". l t
it.
ITt
It ~ I I ~ IIIt
a Nlifl in!.m
LIMZW -£1
j!"2 298
I
I a 11-40;
Lit Xt it44.I Z, J i Fh :m !i, 'i;i titT; p T'" + :i: L%+
I kii, lir 115' ;!:. :!:; , I : . i 1 litHit 1;!I. it': -A :
i;l; HIINJill 1!1 Vj ill L. iotnUM4.
4H-n ti r.- HitVITILH]; UU:HIJ li'll"91 P
Nr 71M.tI P, 41. it F i, ll! In:1i ilT , :J':
U... i:i I ..V7 7AN7: :7,7 1':;
.4 Ha 4itN IF.I 10 ! 11; :.,i: Ti I I4-1 W! L! i , I,,: 1--if . ;;: :-1,i lt! 1: lidi! ;it IlLm P! I Ii L K :T
:--ipl, p ýi I!!. it 41-W::I tiý Ip I:!: N el
i F. i:!Ti:i Wl7 .2" it Un:;it Ili: I i:ýi
ini . IIii:i IAmp : :11iI Hlit 40 1111 li h I
r) C4 164! 11 ;.::1[
it Ifli .It Fit. VA
!,- -:-- I I . it.w. -:v nj
NI . :14;
lit -I
cl, I ..... .... 0
7 U
164 !I'!,] Inj I1::; F lit
.177 "Hit I I I M-1. I 'It C14Nib41 7-7It:J W111ii; IN id: ill!!: I fi ugIlfiIN :: .;_ I ffil... ;.- w W.. i '5! -14. TM U1k ill 1. I- ill -v
P4 W RX1!1!11: .11 1. i; ,:. i it'P. HIM n-ITn. 1! 1 Trrll*.itIT 110 MOM Ili Ir, W : It, iý!: q ... ...if.: UP I H ::I - . N7.:Ulixi
Q) ill OPT Ihi;EJ 1ý illi;, ill, f;;Tr n t'.. --i-T.., I i- d IIPT,ýa Ill. 1, M111. :k . ki.j :, ý; Ill: ill. :: - ---it It 41 p ANt -4. iL NN `T:N IIffliK: i;
c') N"ýT. T.1r, iin C14
in r-N
LrIrrI1-4 -1. dr
U UMn,
-H 77
Hid 0;r4:;I I,
51TRIM.-D Alit;! Li L I
911611 it MIWLV" 1-M
2-29
B. DSG, FABRICATIE. AND DEVELOP AUTOMATIC CONTROL SYSTEM
1. Basic Control Strategy
The automatic control system (shown schematically in Figure 2-25) is composed of A
three interconnected loops: heater head temperature, firing rate (fuel flow),
and airf low. The airf low loop is slaved of f the fuel f low to maintain air/fuel
ratio; the firing rate loop will accept input from either the heater head
temperature loop or an operator set point (manual mode). The air and fuel loops
incorporate proportional integral -derivative control algorithms to stabilize
operation, and have high and low flow rate bounds set at the limits of control-
lable flow through the system.
2. Control System Functions
The function of the combustor controller is to provide: i
* start-up inter loc~king similar to the existing manual control;
0 closed loop air/fuel ratio control (operator set);
0 closed loop fuel flow control based on operator set point (manual
mode) and head temperature (auto mode);
* automatic shutdown due to fuel pressure high or low, blower air
pressure low, exhaust fan failure, fire alarm) emergency stop
(operator), overtemperature, and command from HF Data Acquisition
System (DAS);
0 display: status of system monitors, firing rate, air/fuel ratio,
and mean/max head temperatures;
* controls: firing rate, air/fuel ratio, max/mean head temperatures, :
blower, fuel, and igniter on/off, auto/manual, and emergency stop;
and, I
0 watchdog timer. 1
2-30 I
, -4
-4) 04
00
0 0 , O 4 J
+3 M41 ýa 0
++
P-4
A ".4
0r41
a;*44. oJ.I
"-4
0-3 ri~~0 0-~ --- _ _ 0 _ _ _
In addition, the combustor controller will allow for expansion to include auto-
matic start-up sequencing, control based on mass flow rates, integration with
power control, and transient power/head temperature control.
The combustor controller will also:
0 leave existing manual control packages in place as backups;
0 provide standard connectors for system monitors, relays, valves,
and displays so that conversion from existing control to the new
automatic control can be easily accomplished;
* route all signals through microprocessor control;
0 provide emergency-stop function (button) to interrupt power to
igniter relay, fuel valve/solenoid, and blower valve, and apply
power to blower relay, assuring shutdown in the fail-safe mode; and,
*have tasks to fit controller to test cell to provide- in-lineI
connectors for system monitors, relays, valves/displays, and ther-
mocouples (T/C); additional ovorhead displays for mean and maximum
head. temperatures; and, a new system control panel.4
A block diagram of the combustor controller is presented in Figure 2-26.
3. Control S stem Specification
Table 2-3 presents the control system specifications.
2-32
8 AvTde'iU .zo~u~dO o~uiaiO .o:jwaado
4 Hr 0) 0 -H0 .4-4 0 0 m. -4 0-. 0"A C, 10 4-4 4-4 n P4 0 0. -
0 -4 -~ 0 a- -- 44 r-4 4J ) 0 V4w (W H ý H 00 4-4 COI W. P, 0, *w- I-H
H H0 0 - 0 - 0 0 * P -,I I-CCH~04 W 00 u "0 0) r 0. *v4 t4 11"-=0 0U w 4u 1 H C
;4. 0 = 4.1 (U " - H, ý 00 04- 41 w -aI-. Hr 0 k -4 Ht r 4 H4 0 0. $4 0 0 - 1-a H - co 01.4 I-4 X~ 0 4 0 0 W4- " 0 W .-W V 140 0 w0.0
-H 4 0 -C -4 W 20 A 44 M 44 W P4 0 H
>- (S4 4-i1
0 CD~ a 0 e w
to 00
14 40 -u
0 u4
U) a))
wU 00H H 4
?-I E -4 ON H60 -1 0 0d
1:1
00
E- 000wH
$4 0 01111114 1410H~ w E4-4 H4c
w ILa 0) c0Ia.0 W b 4
w0 H H 1-H 1
s:/iOTp82U~0d 28S 93UTod -jS
2-33
TABLE 2-3
COMBUSTOR CONTROLLER SPECIFICATIONS
I. Input/output - see following pages
II. Monitor to provide:
A. Modification of parameters of running algorithms
B. Line assembler for writing short algorithms
III. Protection to *nclude:
A. Watchdog timer to reset system if CPU fails.
B. Emergency stop to act directly on system as well as CPU; will turn offigniter and fuel, and open air valve.
C. Shutdown due to presently installed system interlocks; being low fuel andhigh pressure, timer/fire alarm, and blower.
D. Shutdown command from HP-DAS
IV. Control
A. Firing raite to be determined by operator in manual mode or by headtemperature versus set point in automatic mode.
B. Airflow to be detteiiined by firing rate and operator set air/fuel ratio.
V. Manual/Auto
A. Manual mode to allow operator start-up of system, and provide closed looptemperature control.
VI. Response/Accura',iy (Goals)
A. Maintain temperature within ±200C (steady-state power).
B. Maintain air/fuel ratio within ±5%.
2-34
TABLE 2-3 (Cont'd)
Input/ Analog/Signal Name Output Digital Range
Head Temperature 1 I A 0 • 40mV2 I A3 I A4 I A5 I A6 I A7 I A8 I A9 I A
10 I A
Ref. Junction Temperature I A 0 l 1OVAir/Fuel Ratio I AFiring Rate I AMaximum Head Temperature I AMean Head Temperature I AAirflow I A 0-8VFuel Flow I A O-8VBlower Off I DBlower On I DIgniter Off I DIgniter on I DFuel Off I DFuel On I DFuel Pressure Low I DFuel Pressure High I DBlower Air Pressure Low I DExhaust Airflow I DTimer/Fire Alarm I DData System I DEmergency Stop I DDisplay Set Points I DBlower Starter Relay 0 D 120 VACIgniter Relay 0 D 120 VACFuel Solenoid 0 D 120 VACMaximum Head Temperature 0 A O-2VMean Head Temperature 0 A O-2VAir Valve Control 0 A 4-2OmAFuel Valve Control 0 A 4-2OmASequence Complete 0 DAuto Mode 0 P
Auto Start ControlStart I D
Mass Flow ControlUpstream Air Temperature I A O-40mVUpstream Air Pressure I A 0-10VUpstream Fuel Temperature I A O-40mVUpstream Fuel Pressure I A O-lOV
2-35
TABLE 2-3 (Cont'd)
Inpi.,tt/ Analog/
Signal Name Output Digital Range
Power ControlOutput Voltage i A
Output Current I A
Battery Voltage I A
Frequency I A
Field Current I A
Cooling Water Temperature I A
Voltage Set Point I A
Overstroke I D
Low Engine Pressure I D
Battery Rate of Charge 0 A
Field Voltage 0 A
Transfer Control Voltage 0 A
Water Pump 0 D
Cooling Fan 0 D
Ouptut Voltage Display 0 A
Frequency Display 0 A
Output Current Display 0 A
S~2-36
TABLE 2-3 (Cont'd)
HP - Data System: (to be added later)
Data system used to monitor engine, auxiliaries, and shutdown in case of over-stroke, overpower, loss of pressure, loss of coolant flow/excessive AT, andquestionable engine dynamics.
Signal Description
Head Temperature 1... 10 - Type K T/C's measuring combustor/head temperatures.T/C's are routed through a reference junction and a single-pole, low-pass pas-sive filter (max signal = 41 mV =10000C).
Reference Junction Temperature - Output of a temperature sensor affixed to theT/C reference junction (sensitivity = lmV/*C).
Air/Fuel Ratio Set Point - 0-10 V signal controlled by operator (specifies de-sired air/fuel ratio).
Firing Rate Set Point - 0-10 V signal controlled by operator (specifies desiredfiring rate (man) or maximum firing rate (auto)).
Max Head Temperature Set Point - 0-10 V signal controlled by operator (specifiestemperature limit for all T/C's; if limit is exceeded, system will shut down).
Mean Head Temperature Set Point - 0-10 V signal controlled by operator (speci-fies desired mean of T/C readings).
Airflow - 0-8 V signal propcrtional to volumetric airflow (8V = 21.9 cfm).
Fuel Flow - 0-8 V signal proportional to volumetric fuel flow (8V = 1.25 cfm).
Max Airflow Set Point - 0-10 V signal (secifies maximum permissable airflow -
protected set point).
Max Fuel Flow Set Point - 0-10 V signal (specific-i maximum permissable fuel flow- protected set point)
Head Temperature Loop PID Control Parameters (3 signals) - 0-10 V signals (spec-ifies constants to be used in temperature loop algorithm - protected setpoints).
The following six signals are active low TTL signals from momentary contactswitches controlled by the operator. Only one signal may be active at a giventime, or all w'.ll be ignored.
Blower Off - Turn combustor air blower off.
Blower On - Turn combustor air blower on.
Igniter Off - Turn igniter off.
Igniter - Turn igniter on.
2-37
TABLE 2-3 (Cont'd)
Fuel Off - Close secondary fuel solenoid valve.
Fuel On- Open secondary fuel solenoid valve.
Fuel Pressure Low - Active low TTL signal indicating fuel supply pressure islow.
Fuel Pressure High - Active low TTL signal indicating fuel supply pressure ishigh. iA
Blower Air Pressure Low - Active low TTL signal indicating outlet air prespurefrom combustor blower is low.
Exhaust Airflow - 120 VAC signal which, when present, indicates test cell ex-haust fan is working.
Timer/Fire Alarm - 120 VAC signal which, when present, indicates 15-minute ex-haust fan timer has timed out, and that the fire alarm is not tripped.
Data System - Active low TTL signal from DAS commanding combustor shutdown.
Emergency Stop - Active low MTL signal controlled by operator, commandingcombustor shutdown.
Display Set Points - Active low TTL signal controlled by operator, commandingmean and max head temperature set points to be displayed on control panel.
Auto - Active low TTL signal commanding switch from manual to automatic mode(operator controlled). i
Manual - Active low TTL signal commanding switch from automatic to manual mode(operator controlled).
Blower Starter Relay - 10 mA current sink controlling solid-state relay (SSR)whose output is 120 VAC, and indicator lamp.
Igniter Relay - lOmA current sink controlling SSR whose output is 120 VAC, andindicator lamp.
Fuel Solenoid Relay - 10 mA cu:rent sink controlling SSR whose output is lZOVAC, and indicator lamp.
Max Head Temperature Display - 0-10 V output proportional to maximum headtemperature.
Mean Head Temperature Display - 0-10 V output proportional to mean head temper-ature.
Air Valve Control - 4-20 mA current output to I/P converter for air controlvalve.
Fuel Valve Control - 4-20 mA current output to I/P converter for fuel controlvalve.
2-38
TABLE 2-3 (Cont'd)
Sequence Complete - 10 mA current sink on when start-up sequence completed andautomatic mode can be entered.
Auto Mode Display - 10 mA current sink on when in automatic mode.
Max Fuel/Air Display - 10 mA sink on when airflow or fuel flow is at maximumallowed.
Overtemp Displa' - 10 mA current sink on when system is shut down due to exces-
sive head temperature.
Fault - 10 mA current sink on when software error is detected.
Air/Fuel Display - 0-.10 V output proportional to actual air/fuel ratio.
Firing Rate Display - 0-10 V output proportional to actual firing rate.
Watchdog Display - 10 mA current sink on when watchdog timer times out.
Timer/Fire Alarm Display - 10 mA current sink on when timer times out or firealarm is activated.
Exhaust Air Off Display- 10 mA current sink on when exhaust air is detected off.
Hi/],o Fuel Pressure Displays - 10 mA current sinks on when fuel pressure is de-tected high or low.
Low Blower Pressure Display - 10 mA current sink on when blower pressure is de-tected low.
2-39
4. System Software
The software will be comprised of five main modules:
1. System Monitor - This package will consist of modifiled versions of
TIBUG and LBLA, and will allow monitoring and dynamic modifica-
tion of all control loop parameters, as well as minor modifica-
tions to other software packages. This routine will execute
during free time between other routines.
2. Initialization - This module will run immediately after a system
reset, will initiate all control parameters, and will start the
system and watchdog timers.
3. Thermocouple Package - This module will take four readings from
each thermocouple,. linearize the results, and check for bad T/C's
or overtemperature. This routine is started by the main control
module (4), and is driven from the end-of -conversion (EOC) inter-
ruzpt of the TIC A/D converter. The module will be run once per
second.
4. Main Control Module - This module will inut all operator
coummands/set points and system status monitors or flow rates,
compute new flow rates, and update valve positions and operator
displays. This routine will run every 0.1 seconds, causing Module
3 to run every 1.0 seconds.
5. Shutdown Module - This routine will run whenever an abnormal
condition demanding immediate shutdown occurs. (Note: This is
not entered during normal, operator-controlled shutdown.)
Software developed for the combustor controller is provided under separate
cover.
2-40
III. EVALUATION OF FPSE INTEGRATED CONTROL SYSTEM
The WPAFB combustor controller was designed to be upgraded in order to provide
other necessary FPSE control functions in addition to the combustor controller
that would eventually lead to stand-alone, unattended operation. These control
functions can be divided into two areas: engine load control, and start/stop
seqli "
A. ENGINE LOAD CONTROL
The engine load control for an engine-generator application must be designed to
control the engine such that the output is maintained within specified toler-
ances of frequency, voltage, and powet output. In addition, the load control
must protect the system from adverse conditions such as load transients (both
increasing and decreasing), momentary overloads, piston/displacer overstroke,
and loss of load.
An uncontrolled FPSE typically responds to load changes by increasing and
decreasing piston strokes; however, changing piston stroke results in changing
voltage, making this unacceptable for an engine-generator system.. Several mech-
anisms by which the engine stroke (and therefore output voltage) can be
controlled as load is changed include:
"* heater temperature;
"* displacer spring damping;
"* displacer spring stiffness;
"• engine presssure; and,
"• engine dead volume.
All of these mechanisms can be used to change the output power for a given stroke
or voltage. Although the mechanisms for effecting the control concepts listed
above have not been defined, some of their characteristics can be identified.
In general, some performance penalty is associated with all control modes, and
the control response to load transients is likely to be a problem with most of
the above control modes. This is a particular problem for free-piston engines
since they have relatively low inherent energy storage when compared to their
output. Therefore, during fast-load transient, the engine may either shut down
I3I!
(during up-load transients) or overstroke (during down-load transients) before
the control system cart provide control. In that case, this particular
engine-generator application requires fast response to transient loads, and it
may be necessary to isolate the engine/ alternator system from the load by either
a flywheel. storage device or batteries that would provide power
absorption/capability during fast transients until the engine power control
system (displacer spring stiffness, pressure, etc.) can respond by changing the
engine operating point to match the new load conditions. For unattended site
operation, this load-following capability is considered necessary, as opposed
to the fixed point operation.
Whatever engine power control is selected for the engine-generator system, the
engine power feedback control loop can be incorporated into the WPAFB control
microprocessor via software changes to provide a complete FPSE control system.
(This would also require additional digital and analog I/0 ports.) The most
challenging task for implementing engine control will be designing and develop-
ing the control actuator into the engine proper.
B. START/STOP SEQUENCING
The task of start/stop sequencing is necessary for "one-button" startup/ shut-
down of an engine system, and shutdow-' due to emergency conditions, which is
ideally suited to a digital control such as the microprocessor-based WPAFB
control. The controller in this case turns on the various subroutines necessary
for engine start-up, and monitors the system to ensure proper response at each
step. A microprocesor -based system can be easily used to evaluate several
alternate start/stop sequences during the control development.
The start sequence for the engine system, shown in Figure 3-1, indicates the
major steps that must be followed during start-up. Both hardware and software
modifications will be required to implement this start sequence into the WPAFB
control; however, the present controller's software can be modified to provide
the combustor start-up (detailed in Figure 3-2), since the I/0 signals have been
incorporated in the present unit.
3-2
Start Combustor
Activate Electronics
Turn On Blower
100%C Head Temp
Wait I Yes
Center P1'• mi, & Displacer as Required
Turn On Alt Field
Motor Engine Piston
Zero Input from Power SupplyWait No I Yes
Turn Off Starter
j Load Capacity Greater than IdleNo e field, turn on field
Wait . comb. blower, turn on blower* cooling system, turn on cooling
Switch from Battery Aux to Engine
Design
Wait Ready for Load e VoltageW e TempNo Yes
Close Load Switch
Figure 3-1 Engine Start-Up
3-3
- - --- -- ._
Turn On Comb. Air BlowerOpen Air Control Valve
I I
Purge Combustion Chamber I
SeI/F ue oOe Lo etnTurn On Igniter
Turn On Fuel
Has Ignition Occurred in 30 SecondsI
No Yes
Turn Of f Igniter
Turn Of f Fuel
Turn Of f Igniter
No Is Flame Still On
Enable A/F Feedback Control
Enable Temp. Feback Control
Figure 3-2 Combustor Start Sequence
3-4
iI-
The stop sequence for the engine, shown in Figure 3-3, indicates the major steps
required for safe engine shutdown. Again, the combustor stop sequence can be
implemented in the present control hardware with the addition of software
changes.
The development of engine power control, including start-up and shutdown, is the *next step to providing FPSE unattended operation.
- 3-5
2¢p
Turn Off Fuel
I 4 Head Temp to Drop Below 1000C
and Purge Comb. Chamber
Turn Of f Motor Power Supply
Wait No iYesTurn off Blower
Ir
Turn of f Alternator Field
Turn Of f Cooling System
Figure 3-3 Engine Stop Sequence
3-6
IV. SYSTEM INTEGRATION
The combustor controller developed and com-.onent -tested (as described inA
Section 2.1) was integrated into the TDE test facility, thus identifying a
number of minor problems that could not be addressed in the component
bench-test. A number of small interactive problems with the controller, test
facility, and engine that were identified and resolved before the start of the
controller test evaluation include:
* Software Problems:J
- no limits were set on derived calculation values; and,
- information was not properly transferred to memory.
* Air/Fuel Valve:
-discrepancies noted between controller and system valve
characteristics ;
-needed to recalibrate valve position versus voltage; and,
-air valves only opened to 80% of full-open position.
* Sensors:
- sensors were inappropriately calibrated;
- error in measured airf low identified; and,
- reset gain.
* Acoustic Resonance:
-- sensors operated with significant resonance;
- needed to damp acoustic noise; and,
-had to decouple combustion blower bypass from system.
The controller was tested in the manual mode after resolution of the above prob-
lems, and controller gains were adjusted to ensure a stable system. A time
constant of 2 to 3 seconds was observed with respect to a step change in firing
rate. Upon successful system integration, the controller was tested in the
ni~tomatic mode to evaluate system operation.
4-1
V. TEST AND EVALUATION
A. SYSTEM TEST OF COMBUSTOR CONTROLLER
The objective of the WPAFB Digital Combustor Controller Tests was to demonstrate
operation of the digital t.ombustor controller, and to evaluate system response
to load changes. Due to time constants within the program, the manual mode of
controller operation was demonstrated with the TDE assembled with a 70%-porosity
Netex regenerator, and the heater head loading was mapped while testing the
performance of the TDE with the regenerator modification. The TDE was then
reassembled in the baseline condition (91%-porosity, 100-mesh .001 wire screen)
during which the start-up and automatic transient tests were conducted.
1. Steady-State Tests
The TDE characterization test matrix is shown in Table 5-1. Curves of total
cycle power and efficiency, heat input to the engine, combustor firing range,
and combustor efficiency are plotted in Figures 5-1 to 5-5. The test procedure
was to hold T/C #12 (heater control temperature) constant at 400, 450, and 5000 C
while using the computer controller in the manual mode to vary the firing rate
as the engine was stroked- to each test point. These tests were necessary to
ensure that the addition of the controller did not affect engine performance or
power control. Based on these steady-state results, operation with the new
controller was judged to be similar to previous TDE operation and, therefore,
acceptable.
2. Transient Tests
The transient test matrix (shown in Table 5-2) was modified to:
"* stay within the operating bounds of the TDE; and,
"* allow continuous transient prints from terminal point to initial
point of each transient.
Figure 5-6 shows the transient steps in controller mean temperature and piston
stroke for the matrix in Table 5-2. Each step change in Figure 5-6 demonstrates
the controller's ability to maintain mean head temperature. Figure 5-7 shows
5-1
TABLE 5-1
STEADY-STATE TEST POINT MATRIX
Heater Head Control Temperature Piston Stroke
4000 C 1.4
1.6
1.8
2.0
2.2
450 0 C 1.4
1.6
1.8
2.0
2.1
2.2
2.3
2.4
500 0 C 1.6
1.8
2.0
2.2
2.4
Notes: 1. The heater head control temperature was the highest reading head T/C
(T/C #12) to be consistent with the ECUT Regenerator Test.
2. Manual mode of combustor control.
5-2
1800
Characterization Test1600 Toontrol
a = 400b - 450c = 500
1400"
1200-
-----
1000 .. I
800--
600- a
a a400 f
200
0.0 .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6
Piston Stroke (cm)
Figure 5-1 System Power Versus Piston Stroke
5-3
37.5
Characterization Test
35.0 Tcontrol
32.5 a - 400b - 450c - 500
30.0
27.5
25.0
X* 22.5b C
S• ¢ •C
20.0
15.0.i
12.5
10.0
7.5
5.0
2.5
0 C |1 IiI I I I ~ .. I i i !
.0 .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6
Piston Stroke (Cm)
Figure 5-2 Total Efficiency Versus Piston Stroke
5-4
II 1
10I0
Characterization Test
8 Tcontrol
a = 400b = 450c = 500
33
6b
aI
o /H
w 5 - • -
2
1 I
0 .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6
Piston Stroke (cm)
Figure 5-3 Heat Load Versus Piston Stroke
5-5
Ici
Characterization Test
Tcontrol
8 a -400*
b - 45O*
c - 5000
77
6.
b
b
aa
31
0 .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6
Piston Stroke (cm)
Figure 5-4 System Firing Rate Versus Piston Stroke
5-6
10
9 Characterizat-ion Test
C8 C C
C
C i.61
c 5
144QJJ
5--4
Heater Control Tempa = 400'Cb = 4501C
C = 5000
3
2
6itI0 .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 I
Piston Stroke (cm)
Figure 5-5 Combustor System Efficiency Versus Piston Stroke 15-7
TABLE 5-2
TRANSIENT TEST POINT MATRIX
Transient Step
Controller Mean Piston Stroke Temperature Stroke
Temperature (OC) (cm) ( 0 C) (cm)
200 Start-up Start-up
400 2.0 +.2
400 2.2 +50
450 2.2 -50
400 2.2 -.2
400 2.0 +50
450 2.0 +.2
450 2.2 +50
450 2.2 -50
400 2.2 -. 2
Notes: 1. Automatic mode of combustor control
5-8
2.4
2.2 A,"
2.0 L 4i. y.:. -,
-~1.80
& 1.6
P4 Engine Stat
1.4
1.2 ]
1.0
4501
400
350
S30004
S250H0
S200.i
0-I
'•150
U
10 20 30 40 50 60 70 80 90 16oTime (min.)
Figure 5-6 Controller Mean Temp. and PistonStroke Versus Time
5-9
A= Firing Rate (kW)
1M C- Piston Stroke
7
4- F
012030 40 50 60 70 80 90 100
Time (min.)
Figure 5-7 Various Parameters Versus Time
5-10
combustor system responses for these transient steps (see in Figure 5-8). The
most radical change is in the firing rate as the controller opens the fuel valve
to fully open for demands in increased'mean temperature, and closes the fuel
valve to 2 kW for decreased mean temperature demands. More severe transient
steps were conducted during the gaxn-setting phase of the controller checkout
tests. An engine start-up transient to steady state is illustrated by the burn-
er system responses shown in Figure 5-8. The responses from a mean controller
temperature step of 500 to ^'0OC are shown in Figure 5-9, and an A/F step of 30:1
to 20:1 is shown in Figure 5-10.
3. Conclusions
The combustor controller maintained heater head temperature in response to load
changes of the system, and performed as expected by:
* automatically maintaining heater head temperature with respect to
air/fuel rates and/or engine load changes;
"0 providing start-up interlocking with the TDE facility;
* providing close-loop control of air/fuel ratio and temperature mean
max with operator set capability;
"displaying key operating parameters;
" providing automatic shutdown; and,
* providing expansion capability to interface with power control
system.
The development of the combustor controller is the first phase that will lead to
unattended FPSE operation. (A picture of the combustor controller as installed
in the MTI TDE/FPSE test facility is shown in Figure 1-5, input/output signals
to the combustor controller are shown in Figure 1-4, and the interior of the
control is shown in Figure 1-3). The main features of the operating controller
are:
5-11
Start-lip Transient
A Firitig RateB -A/F*le-1C Piston StrokeD -Controller Mean Temp*le-2E -Preheat Tenaj*le-2
IC F -Max Fin Temp*le-2 Exhaust TempG Head Exit Teup*le-2TepH -Ex. Temp*1e-l
90.
Max Fin Temp.
Firing Rate
Head Gas Exit Temp.cd
Controller Mean Temp
D.Preheat Temp.
,,-Piston Stroke
E ngine Start with~6 1Load
010 15 20 25 30 35 40 45 50 55 60
Time (mmn.)
Figure 5-8 Various Parameters Versus Time
5-12
500c to 250c Transient
A Firing Rate (kW)'
B A/F*le-IC Piston Stroke
1. Controller Mean Temp*le-2
E = Preheat Temp*le~ 2
F = Max Fin Temp*le-2
G = Head Exit Temp*le-2
9 H - Ex. Tenip*le-2
8C
S6
F 1
E Max Fin Temp.
Preheat Temp.
• • .,aA! F
SController Mean Temp.
"C R • ead Gas Exit Temp.
_D Firing Rate
piston Stroke
A Exhaust Temp.
H
0 1 2 - 4 6 7 B 9 O ll 12 13 14
'rime (min.)
Figure 5-9 Various parameters Versus Time
5-13
:I A/F Transient
A Firing Rate (kW)B A/F*1e-I
10i C, Piston StrokeD O Controller Mean Tenp*ie-2
E Preheat Temp*le-2F - Max Fin Temp*le-2
SG - Head Exit Temp*le-2H - Ex. Temp*le-2
8
7
Max Fin Temp.
Firing Rate5
Head Gas Exit Temp.Controller Mean Temp.
E "Preheat Temp.
3 IPiston Stroke
2 C
Exhaust Temp.
00 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Time (min.)
Figure 5-10 Various Parameters Versus Time
5-14
"* start-up interlocking with facilities,
"* closed loop air/fuel ratio control with operator set capability;
"* closed loop airflow and fuel flow control;
"* automatic control of heater head temperatures;
"* display of key operating parameter-s;
"* automatic shutdown in the event of key failures in the system; and,
"* expansion capability to interface with power control system.
The range over which the controller has been tested includes:
*firing rate 2 Lo 10 kW
*temperature 200 to 5000C
o air/fuel ratios 15:1 to 35:1
* set temperature error ±10'C (versus goal of ±20 0C)
The above limitations in operating parameters are operating limits of the TI)E
and test facility, not a limitation of the controller. The controller performed
as expected; other comments on the entire system performance are:
*the largest thermal inertia in the burner system is in the recupera-
tor response;
"* during light-off, the air valve is almost closed while the fuel
valve is open (results in a very rich light-off of the order of 2-5
to 1; the burner does light, and soon after ignition, the air valve
opens to the proper precet air/fuel ratio); and,
"* an engine start from 2 00'C mean temperature was accomplished in 3 to
4 minutes with no difficulties during the transient.
5-15
VI. LIQUID-FUEL COMBUSTOR/EXTERNAL HEAT SYSTEM
EVALUATION AND DESIGN
A. EVALUATION OF MONOLITHIC HEATER HEAD
In addition to the evaluation of the combustor controller, an analytical evalu-
ation of the external combustion systemn%, with emphasis on the monolithic heater
head, was performed to assess the adequacy of the system, and its potential. for
incorporating various military logistic fuels.
1. Performance of TDE Heater Head
The TDE utilizes a monolithic heater head (shown in Figure 6-1) in lieu of the
traditional tubed design. The advantages of the monolithic heater head include
eti.mination of the thin-wall tubes, elimination of the structure tube-to-headbrazed or welded joints, and potential for a long-lived heater head.
The TDE heater head as fabricated exhibits an axial temperature distribution**
(shown in Figure 6-2), as opposed to uniform temperature distribution, that
limits its capability for peak temperatures and, hence, higler performance. The
TDE heater head was the first monolithic heater head developed. Improvements in
analytical techniques and imperical data from the head have led to improved
heater head designs that will be built and tested as part of ITI's continuing
FPSE Development Program. Subsequent testing of the EM monolithic heater head
substantiates its much improved temperature distribution over that of the TDE
head.
The problem with the TDE heater head is the external and internal leading fin
geometry. The outside fins transfer too much heat to the base metal, while the
internal fins are not capable of transferring the same amount of heat to the
working gas. Hence, the leading base-metal temperature is too high, limiting
temperature based on structural constraints. The mean heater head temperature
*consists of heater head, preheater, combustor, fuel nozzle, air/fuel control,
fuel supply system**discussed in Section IIA, and shown in Figure 2-8
6-1
TDE-\ r Natural Gas Heatvr
F.ic I :ozzIc Recuperator Head
xI
AirOutlet (a) TDE Gas-Fired Combustion System
Dome Fin Entrance Dome Dome Fin Entrance
Gap Gap
Zone B Zone B
Gap Gap
Zone C Zone C
(b) TDE Finned Hqater Head
i ~8131181
Flgure 6-1(a) IDE Gas-Fired Combustion System(b) ThE Finned Heaacer Head
6-2
700 --4.98 kW
9head
600 "
A(13) I
(22) A(15)
"• / • ~~B(23) 5),
A(9) A 178-400A(8
B I
300
A & B Represent T/C atTwo Circumferential Locations
200
100
0 0.24 0.50 0.75 1.0 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00Thermocouple Axial Location from Fin Leading Edge (in.)
Figure 6-2 Heater Head Temperature Axial Distribution
813174
6-3
is considerably less than this peak heater head temperature. This can be seen
in Figure 6-3 where, as stroke is increased (power increased), the leading edge
fin temperature remains almost constant, while the dome (protected by a ceramic
cup and indicative of expansion space temperature) continues to decrease.
Development of a monolithic heater head that provides uniform temperatures is
necessary so that maximum potential of the engine system with temperature can be
realized. This has been accomplished with the development and test of the EM ,
monolithic heater head. I
2. Suitability of Monolithic Heater Head for Air Force Logistic Fuels
There is no known limitation for use of the monolithic heater head with USAF Ilcgistic fuels. The only concern would be high-temperature corrosion of the
heater head; however, for the materials and peak temperature limits that would
be selected for FPSE monolithic heater heads (i.e,, Inconel 713 - 14000F,
Inconel 617 - 1200 0 F, and SS 316 - 11000 F), no significant corrosion would take
place. In addition, MTI reviewed the entire external combustion system in a
separate study, concluding teat it is feasible to utilize gasoline, jet fuel,
and diesel fuel in a single FPSE external-combustor-system design. The only
disadvantage is that as sulfur content of the fuel increases (gasoline 4 jet
fuel - diesel fuel), the amount of sulfuric acid in the exhaust productsincreases proportionately. As long as the acid remains in a gaseous state, no
problem exists; however, if the exhaust is cooled below the dew point (-30 0 0F),
a corrosive condensate is introduced. If either the average exhaust gas temper-
ature in the preheater or the local gas temperature in contact with a cool
preheater wall are below 300 0 F, condensation occurs.
United Stirling of Sweden (USAB)* has experienced coriosion with the P-40 stain-
less steel preheater while using diesel fuel, but none with unleaded gasoline. 9Prevention of this would require reducing the preheater effectiveness (higher
exhaust temperature and lower cycle efficiency), or using a material impervious
to sulfuric acid (possibly ceramics); therefore, the conclusion is that not only
is the monolithic heater head acceptable for multifuel use, but the entire1
"*major subcontractor to MTI's Automotive Stirling Engine (ASE) Development
Program
6-4 i
600 -
500
, ti
400
Heater Head Zone.AHeater Head Zone B
:300 Heater Head Zone C _am Dome
amI
100 ,
TDE Finned Heater Head
0 0 1 1 1 1..40.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2-. 0 2.2 2.4
Piston Stroke (cm)
Figure 6-3 Average Heater Head Temperature Versus Piston Stroke
6-5
r -M
combustor system can be designed to be satisfactorily operated with Air Forcelogist ic fuels.
B. EMISSIONS OF THE TDE EXTERNAL HEAT SYSTEM
Emissions of the TDE external heat system were taken and evaluated as part of
this program. Gaseous emissions (NOx, CO, and HC) were measured on November 3,
1981. A total of nine data points were taken using a matrix of air/fuel ratios
(35, 30, and 20 by volume) and maximum heater head temperatures (400, 450, and
.5000 C) burning natural-gas fuel. The results are presented in Figures 6-4 and
6-5 on a volumetric and mass (Emissions Index) basis, respectively, as a func-
tion of Lambda (X), which is defined as air/fuel divided by the stoichiometric
(theoretical for complete combustion without excess air) air/fuel.
As illustrated in Figures 6-4 and 6-5, none of the gaseous emissions show any
dependence on control temperature. The apparent scatter in CO is most likely
due to the fact that the high-range CO analyzer (0-5000 PPM) had to be used
because of the unavailability of the low CO instrument. (0-500 PPM). For the
formation of pollutants, relevant TDE parameters and their effects are:
Engine Parameter Influences Effects
Preheat Temperature Flame Temperature NOR, CO, HC
Air/Fuel Ratio Flame Temperature NOx, CO, HC
Airflow Residence Time NOx, CO, HC
Heater Head Temperature Catalytic Reaction CO
Heater Head Heat Load Reaction Rate NO, CO
As maximum heater head control temperature is varied, preheat temperature, heat-
er load (engine output),. and mean head temperaturo vary, as illustrated in
Figures 6-6 to 6-8; however, these variations are not very large. A larger
variation in the temperature may have had a more noticeable influence on emis-
sions. Similarly, heater head load (4.1-4.7 kW) and mean temperature
(300-375 0 C) variations are small. A larger variation in heater head load can
also be expected to influence emissions. For the test conducted, the dominant
Iieffect on emissions is due to variations in air/fuel and airf low.
6-6
80 O3 CO
70 dD
60
50 C3
0S40 -
ClC3
0oo
(n 30 -0NO
Tcontrol Note: NO and CO dry sample
20 0 400 HC ref. to methane-wet sample
0 450 Sampling upstreav of preheater
500 110 -
0
2.0 2.5 3.0 3.5 4.0
Lambda nu X 1 (A/F)/(A/F)s
Figure 6-4 TDE Emissions - Lambda Versus Gaseous Emissions
(Natural-Gas Fuel) 822844
6-7
5 -
4 -
3 -I1
NO0 NT control !
132 400 Note: NO and CO dry samples
0 450 HC ref. to methane-wet sampleo• 500 Sampling upstream of preheater
0c
'10 1
C,
"-4
0A 3
2
ij
1 H
2.0 2.5 3.0 3.5 4.0
Lambda • X n- (A/F)/(A/F)a
Figure 6-5 TDE Emissions - Lambda Versus Emissions Index
(Natural-Gas Fuel)822802
6-8
90
TCont rolL 4 8 0 4 0 oQ)
450
70 500
410
400 500
3 9 0oqJ,380'Q) 3701
a4 360
I120 T400
500
12
110 450
14
(n 100400
90
2030
40Air/Fuel Ratio
Figure6-6 Ehaust Temp/Prelleat Temp/Burner riVersus Air/Fuel Rati'O
6-9
450
.T • 0control
050O 400
450
.• 350
400
S300-
10[.. 6506 500
S600 - O--' " m -O 450
"-4
• 550 400
_ 500
. 4.4 - 450.-I.
S4.2-w 400
= 4.0-
I I p I
20 30 40
Air/Fuel Ratio
Figure 6-7 Heater Heat Load/Max Fin Temp/Head ExitGas Temp Versus Air/Fuel Ratio
822711
6-10
T400 cont rol
Q A 5 0 0
3450
105
5Li 001
10'
14450
3
2 20
01 2030 40
Air/Fuel Ratio
Figure 6-8 Area Weighted Mean Head Temperature
and Firing Rate Versus Air/Fuel Ratio 822786
6-11
From Figures 6-4 and 6-5, a strong affect of air/fuel (X) on CO and NO is noted,xwhile H1C is essentialiy constant. As X increases, flame temperature decreases.
Concurrently, airflow increases (fuel flow, i.e., "firing rate" in Figure 6-8 is
nearly constant) and the residence time at that flame temperature decreases.
The effect of both is to decrease NO - exponentially in the case of flamextemperature, and linearly due to residence time. CO, whose oxidation to CO2 is
retarded by decreased flame temperature and residence time, exhibits the I
opposite trend. The extremely low and flat HC levels indicate a more than ample
combustor reaction volume.
Overall emissions levels are extremely low in the case of NO and HC, but are Hx
reasonable for CO. Calculating a combustion efficiency based on the highest
emissions levels of CO (80 PPM) and HC (5 PPM) yields an efficiency equal to99.9%. It must be noted that the TDE is operated at extremely high X's (excess Iair), as compared to the General Motors GPU-3 Stirling engine (1.4-2.7) and
MTI's ASE, P-40, or Mod 1 (1.15-1.4) engines.
Although beneficial to NOx emissions, a penalty is paid in external heat system I("burner") efficiency, as shown in Figure 6-6. If the TDF were operated at a X
equal to 1.8, an efficiency of 90% is possible. This would, however, incur a
penalty in NO emissions (Figure 6-4). The following conclusions can be made:x
"0 small variations in preheat temperature, heater head load, and mean
head temperature did not affect the emissions levels;
* as combustion becomes leaner (increasing X), NO decreases and CO ix
increases, as expected;
* emissions levels are extremely low, i.e., NO was less than 100 PPM,x
and CO and HC indicate a high combustion efficiency;
* the TDE data compares favorably to published emissions of the GPU-3
buring diesel fuel; and, 1
* the gas analyzer CO2 measurements usfd to calculate X agree reason-
ably well with those obtained with the gas chromatograph and the
engine cell airflow/fuel flow.
6-12
C. GPU-3 DIESEL-COMBusTOR TEST AND BASELINE EVALUATION
As a prelude to the design of a liquid-fuel combustion system that will inte-
grate with the MTI FPSE designated Engineering Model, the combustor and fuel
nozzle of the GPU-3 were tested and evaluated. The 10-hp GPU-3 was designed to
operate on diesel fuel, lending its combustor performance information as a valu-
able baseline. The GPU-3 combustor and fuel nozzle (Figures 6-9 and 6-10) were
supplied by NASA/MERADCOM.
The combustor cup has three air entry points:
"* eight slots or louvers located on the conical dome; i
"* six tubes located on the cylixdrical portion of the cup; and,
"* an annulus, formed by the cup and liner.
The slots serve a dual purpose: 1) to cool the dome; and, 2) to import swirl to
the flow, creating a recirculatory region inside the cup for mixing and flame
stabilization. The remainder of primary combustion air enters through the
tubes, Air that flows in the annulus is used to cool the downstream cylindrical
edge of the cup, thus completing the combustion process. The water-cooled fuel
nozzle uses on external supply of pressurized air for atomization, and contains
an igniter located in close proximity to the nozzle exit. The face of the nozzle
also contains eight slots that are fed with a small amount of air via feed holes
upstream of the combustor cup. The air corotates with that entering through the
cup's dome slots, serving to purge the igniter and nozzle.
Prior to fired testing, the fuel nozzle was sprayed with diesel fuel and found
to have an asymmetric pattern that was corrected by replacing the spin plug and
cup. Atomizing air pressure was varied between 1.0 and 1.5 psig, with no affect
on the diesel-fuel pressure versus flow characteristic or spray quality,
Evaluation of the GPU-3 hardware was performed in the MTI Free-Piston Combustor
Rig (Figure 6-11), which uses an electric heater to supply preheated combustion
air to burn diesel fuel. A portable diesel-fuel supply, control, and measure-
ment system was designed and fabricated. Sheet-metal modifications were made to
adapt the GPU-3 combustor and fuel no.zle to the rig. The GPU-3 combustor liner
(Figure 6-9) was replaced with a 3.66-inch I.D., 10-inch long cylinder. The
6 - 1 3
'
kI
Figureto 6ome CooToPiceBrnrnseml
~ Ignitr Sltip
Prima i
L.i
Figure~~~~~~ 6-9 GM w-ic unrAsml
h" +A
Fiur 6-10e GR FulNzl ndIntrUi
6-14
-t
**' t.
I.
a:
CC.4-)'A
0-C) N
S.-LL
'-0
C)5-=
LL
I______________________________________________________________________
I4 I
V j�a.AW.vv.A�4t .4 ,. ¾
4 U'* <'
V¾
6-1 5
_ a
I.D. is the same as the engine liner to preserve the same cooling annulus flow.
Due to the arrangement of the rig, the fuel nozzle and combustor were mounted
the motor-driven engine power-supply system was mounted on the rig.
Tecombustor cup and liner were instrumented with thermocouples (three on the
cup and four on the liner). Locations of the thermocouples are indicated in '
Figure 6-9 and in Table 6-1. The liner thermocouple locations are relative to
the leading (upstream) edge. Rig instrumentation included flow measurements of
air, atomizing air and fuel, air and fuel temperature, and combustor pressure
drop. A bare-wire thermocouple mounted on a traversing rod was usud to give an
indication of axial-gas temperature profile within the combustor. Test condi-
tions for the evaluation were determined from Table 6-1 (Reference 3), and input
from NASA (Reference 4). Due to a limitation in airflow capacity, full-powerI
GPU-3 condition could not be duplicated (24-kW versus 30-kW firing rate). The
steady-state test results are given in Table 6-2 and Figures 6-12 through 6-19.
The conclusion is that both combustor cup and liner surface temperatures are too
high for long life. Both pieces achieved temperatures in excess of 19000F.
Metallic pieces, even if made of high-quality stainless steel, would probably
not survive for more than 100-200 hours. A second conclusion is that combustor
pressure drop is too low (AP/P < 1%) and/or atomization is inadequate to mix
fuel for completing the combustion reaction in a reasonably conservative volume
(15 kW/ liter) . At higher fuel flows and A/F !• 25*, combustion was not completed
within the liner volume, i.e., a luminous yellow flame with burning carbon
particles or droplets was visible above the exit.
The measurements of combustor cup temperatures (Figures 6-12 to 6-14) indicate
that the hottest region is in the cooling annulus, and that surface temperature
is proporticnal to firing rate and inversely proportional to air/fuel ratio.
Although the annulus is cooled, combustion gas temperature is higher than in the
dome, causing high wall temperatures. This was confirmed by an examination of a
used combustor cup where the most severe oxidation was observed on the trailing :edge, i.e., the portion of the cup located in the annulus. The increase in
*GPU-3 nominal A/F =25
6-16
Lin U\CNC 0 0 CD rl. 0 C
co C
0' 00L\ h 0 -
4~ N N N N 04 N
JJ 00 m r--I0 " C . 0IC~~ 0 0 -. U en to -.
cc4 0 2 4 0 O ' \ 2 N 00 - . . N N 0 A r
LL-I
LU
(D ) 0
0.. C\ ONC N n N N 0 '
(3 00 -00
LAO
0 -
N -J6-17 '
M. m f- 00 OD N N 4 4 C 0 0 C0Ou,0 m'. 0 \.0 '.0 WS 0 - 0
0 0 ?0 000n 0 00n0 0 0
000
0~C 0 oA 0'.0 (7 U 0 en C- C 0 0 rn 0 4 C
\O UNl O\~ zr 0 n 4 L - 0 Ln I I I
N ~ ~ ~ ~ ~ ~ ~ ,ý -Lr N ..0 - - C OC- \I- C-C-C- -C-C-r--C-
\N CCC 0.0 -'1 '0 '0\ Z) .7 CO CO Z') tA r- 4)
~c. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ .0- Q)a) C ) 6=;I
r-C. j N - ON C-- N C-- C, 3 CO w 4 %D'.0 : -LA N 0 .0 CO C- 0 N CLC~j-T r- C- -- r-\ \-- 7O z'C- 4 L .0C C C 0
- ~ ~ ~ ~ ~ ~ ~ 4 __________________ L. .C-. O. 0 0
H COO CO T \40 m~ In 0 44r -y COO C .. O w-- 0 0 C-- \J'0 4r '0 C") 4r 00 m N 41 4 1J4 1 t- LLN 0' mA mC z, 7 4 LA LA n N m~ 4 en ~ I)C O
4l C"Z en _ '0\ r- 0O LA\ N~ _4 N CO co Q0- . 0.nC U\ MC' 0 XA CO 0 MC C-- t'- - 4r r-- '0N mC) C CO CO COm ' C O a\ C-(- LA '. 0 '0 L-L LL t-L t-
C-~~~ oo-- C- - - - -C C C 00000000AJ-J ) 444 " 4C 44444)
H - 0 N N 0' - 4 CO N 0 0 .-ON C--CCen CO '0 LA'% '\ 0 mC' \0 0 C- 4- 41EE E E E E eE
-,00 000o 0 C, C
N 0o LA 4C' - m L r'- Z U,0 N 4m XI NNN' N-Q-3~ - - -C-C-C- -C--C-C-C-
> ~0 LA ULA LAý 0 0 0> 0 0 0 LAO% 0CV \0 (* CO N CO Z CO CO mC' ; - N - \.0
C- AC NCD)C,
0
VI) .0 fn LAl ULA LACO go uA 0 4 -LA LA C- 0 '.0 0
LA '0 LA 0 '0 0 0 en) co C- t-- 4T 0. 4l CO LZC 4 U Co - r- N1 No Nn Nm -N C
__ __ __ __ __ __ __ __ __ _ V)
CnC
U0.. 0 0) LA .- LA L ' 0 \0 '0 0\ 01 -.CO4 V N 0ne,~ ) m" N N N\ ff) N .- N 0
0)40) h0- 4-)0 0 0 0 0 0 0 0 0 0ý 0 0 0ý L . -0.
___ ___ __ ___ __ ___ __ ___ __ )c 4141J m
00 mVa)~ 0
LA CO 4 4 4 4 0 LA 4 0 JcC/ 4..)
.6 N co C- j N N N N~ N~ r- 4T :I _ 4 *-.-~ rn n
co 00 0 " 00(V N 0 CO 00 . 07 C 0 C\0h' -J-V-)L " 11-cc C' N C- CC- 3 N7 NV)- V1)1)
-E C- n7 C- .1. c n3 3Oo 0 0 0 0 0 0 aE xn--...fl
LI)-~~~~~~ E N EC C A ' - O ' 0 C C
Q)0- C- C- C- 111111
CV C- C-j
6-18
1900
1800 - T21 roý
1.700 -4 T220.0 T23 C
1600 -
1500
1400
1300
1200A/F 20 [
0 2 4 6 8 10 12 14 16 18 2'0 22 24 26
14) 1900
0
4T221700 - T23 '00
1600
1500
130-A/F 25 r1200
0 2 4 6 8 10 12 14 16 18 20 22 24 26
Firing Rate kW
822817 1Figure 6-12 GPtJ-3 Combustor Evaluation(Diesel Fuel) I
6-19
1900-
1800- T T21
1700- 4 T23 0,
1600 -
1500
1400- ea 1~1300-
P A/F 301200T
SI I I I I .l l I I ,• I _I t
S0 2 4 6 8 10 12 14 16 18 20 22 24 26co
S1900
1800 -o T21Cu • T22 o•
1700 T23
1600 -
1500 -
1400- ?IV S I
1300A/F "35
1200
0 2 4 6 8 10 12 14 16 18 20 22 24 26
Firing Ra te 'I, kW
Figure 6-13 GPU-3 Combubt .valuotion(Diesel Fuel) 822832
6-20
1900
Al F
0 20o 25
1800 - 30
0 35
C)401700
it 1600
S1500
U 1400
0
0o 1300
1200
1100
1000 -
I I I I 1 10 2 4 6 8 10 12 14 16 18 20 22 24 26
Firing Rate " kW
Figure 6-14 GPU-3 Combustor Evaluation (Combustor CupTemperature Eetween Swirl Slots) 823257
(Diesel. Fuel)
6-21
1900
1800
1700
1600
1i.,00
1400
1300 kW TpH %0 F•24.1 1380
1200 4 15.4 1230
P 1100 A/F 20 0 10.3 1110
1• 000,
S0 1 2 3 4 5 6 7 8
0.
$4 1900
1800
1700
1600
1500
1400
1300 -1kW Tp OF
1200 15.0 1275v20.5 1380
SA/F 251100 - 0 7.3 1075
1000 -
II .1I I I , I I
0 1 2 3 4 5 6 7 8
Axial Distance " Inch
Figure 6-15 GPU-3 Combustor Evaluation(Diesel Fuel)
6-22
1900
1800
1700
1600
1500 k l
1400A1300 W 415
1200 A 30
0 6.8 1060
11000 ....
t)0
- 900
AJ 1800
1. 700
160
1600
13 0 2 15, 41225
110
Combustor Liner Temp @ 2.875"
1900
0 20
o3 25•h 30
1800 30
0 40
1700
1600
1500
/
1400
0 2 4 6 8 10 12 14 16 18 20 22 24
Axial Distance " Inch
822829
Figure 6-17 GPU-3 Combustor Evaluation(Diesel Fuel)
6-24
"00:
N 4- 01 3z :3
Ln~S 0 (0-r
0i p.r3U4
'4400
ta
CAj .-H 0)
-4-
Ln Ln CD00
owa 69 0Olnuo 0aoiou
6-25
A/F..70
0 200 25 0
A 30
0 35
.60 0 40
0O4,
.50
SNe .40
oPLIi
.30
.20
.10 0.350
,...I I I L I i I
6.050 0.100 0.150 0.200 0.250 0.300 0.350
822801
Figure 6-19 GPU-3 Combustor EvaluationCombustor Pressure Drop
(D-iesel Fuel)
6-26
surface temperature with firing rate and decreasing air/fuel ratio was due to
higher combustor loading and flame temperature, respectively.
Combustor liner temperature is illustrated in Figures 6-15 to 6-17. The highest
indicated temperature is at 2.875 inches, although the actual peak is probably
between 0.375 and 2.875 inches. Once again, the trend is for surface temper-
ature to increase with firing rate and decrease with air/fuel ratio.
A typical plot of the traversing thermocouple for test point 6 is shown in IFigure 6-18. Although the absolute temperature is not accurate due to the large
radiation error, the trend in gas temperature as a function of axial distance is
indicated. The gas temperatures are consistent with the liner surface temper- t
atures of Figures 6-15 through 6-17. Figure 6-19 shows % combustion drop versus
corrected mass flow parameter. An increase in AP/P to the range of 2-4% would be
benefical to both combustion volume requirements and mixing.
Additional testing was conducted to determ.ine ignition/blowout limits. In the
former case, ignition was unsuccessfully attempted using the GPU-3 to the
inverted position of the igniter, which allowed fuel to wet the igniter and
prevent sparking. NASA experience indicates that ignition is easily achieved
with the combustion system in its normal position. Rather than extensively
modifying the rig to mount the test hardware in its designed orientation, aI
propane torch was used for ignition. Blowout testing yielded air/fuel ratios in
excess of 100*, indicating good stability.
D. MgUID-FUEL COMBUSTOR CUP AND NOZZLE DESIGN
The design of an EM-compatible liquid-fuel combustor and fuel nozzle is shown in
Figure 6-20. The combustor cup is similar to that used on the TDE and EM
natural -gas- fired engines (Figure 6-21), and represents a scaled-down version
of the GMR-GPU-3 diesel-fueled combustor shown previously in Figure 6-9. The
material is 310 stainless steel; however, based on the GPU-3 evaluation, this
material will eventually be changed to ceramic because of the estimated 190COF
or higher surface temperatures. The high-surface temperatures are the result of
*A/F =114 at 600OF combustor inlet temperature
6-27
F-jel Nozzle
Combustor Cup
"0 C- 1 : --- • I ' iner
Ceramic Insulation ~ ± '
Ceramic Flame Shield _ I
MonolithicHeater Head
Ceramic Fin Stuffer .
Figure 6-20 Liquid-Fueled Combustor and Fuel Nozzle
6-28 8228S0 *1
0.250 Diameter Tube
6 Equally Spaced
A
300 Louver is
Punchedoutward
O.D.,
1.690 0.073 1
1.50 TYP
-.7D Louver DetailI.D.
2 x SizeTYP 6 Places
Figure 6-21 TDE Combustor Cup
6-29 L
a 14500F combustion inlet air temperature and an increased f lame temperature/
luminosity of liquid fuel (as opposed to natural-gas). In the former case, the
need to maximize~ external heat system efficiency by recouping energy from the Icombustion gases exiting the heater head results in high air temparatures into
the combustor, reducing the ability to convectively cool the cup walls. The
latter condition results from soot particles in liquid flames that give a char-
acteristic yellow, more luminous flame, thereby increasing radiant loading and,A
because the carbon/hydrogen ratio is -1/2 that of natural gas, leading to higher
flame t empe ratux a/ convective heat transfer to the combustor walls. L
The combustor cup utilizes louvers on the cone portion of the cup for flame lstabilization, cooling, and as a source of reaction air. The remaining
combustion reaction air enters through six tubes. A seventh tube is utilized
for the igniter. The remaining air convectively cools the cup by flowing
through the annular gap between the cup and liner. Pressure drop is under 2%,
but will probably be increased to improve mixing and reduce combustor volume.
Fuel nozzle requirements of the liquid-fuel combustor pose some unique require-ments. Maximum fuel flow is '-0.4 gallons/hour (l5-kW firing rate), falling
within the range of domestic oil burner nozzles. Simple mechanical atomizers
provide a turndown ratio of only 2:5 (inadequate for the 10:15 needed). A more
sophisticated dual-orifice or piloted air-blast nozzle, typical of aircraft gas
turbines, will provide the necessary turndown, but they are no)t feasible in the
low flow range required because of the plugging tendency of small orifices
(characteristic of these nozzles). It is possible to use an air-atomizing
(air-assist) nozzle that utilizes an external supply of pressurized air for
atomization. This type of nozzle has relatively large fuel passages and good
turndown, and is, in fact, used on the GPU-3 (30-kW firing rate). The disadvan-
tages of an air-atomized nozzle include the requirement to provide/drive an air
compressor, and the reduction in external heat system efficiency. The latter
condition results from cold atomizing air bypassing the preheater, and is most
notable at low-power conditions.
Currently, the nozzle selected for experimental evaluation, shown in Figure 6-22 i
(manufactured by Sono-Tek Corporation in titanium or aluminum in two flow
ranges: 0-1 and 0-3 GPH), uses two piezoelectric crystals to sonically atomize
fuel. The advantage. of this concept are wide turndown at low flows, extremely f
6-30
1A
SPECIFICATIONSSC STC STC STC STC STC STC STC
Model Number 01-A 3-A_ 05-A 0W-A{ 015-A 016-A 020-A 021-A
c=•=•ty (ON)-gph 0-I i 0-1 0-3 0-3 0-3 0-1 0-1 0-1
Requrlmenbt'-wete10 6 7 10 8 7 11 7
Spray She" Cylinder cylinder cylinder cylinder 600 60 80, 80Ocone cone cone cone
Mda dropillameter (o.)-
microms 17 17 22 22 22 17 17 17
Nomaze Matedall Ti At Al Ti Al Al Ti Al
Si (Inches) 1I x 2% x 3% x 2N x 31/2 x 2V x 2'" x 21 xI I D I114 D 114 0 I%/ 0 10D 1 0 10E
Wet (ox-)1.7 1.5 3.8 42 9 1.3 1.4 1,3
'Power is supplied by a single transistor oscillator supplied with the nozzle The oscillator can beadapted to either AC line current or low voltage DC operation.
'The Nozzle Material reters to the two metal cylinders which sandwich the piezoelectric crystals.
SONO-TEK Sono-Tek Corporation
ULTRASONIC 313 Main MailPoughkeepsie. NY 12601
ATOMIZING (914) 471-6090040ZZLES
I.iginrt, 0-22 Sono-lek I.Il Nozzi,
,-4I
fine droplet size, minimal electric power, and a 3/32-inch diameter passage that
will not plug. A unique requirement for using the Sono-Tek nozzle is that
mounting locations are limited to the fuel supply tube connection or nodal
points; otherwise, atomization is adversely affected. Figure 6-23 illustrates
the present mounting/interfacing arrangement design of the titanium EM ultra-
sonic fuel nozzle and combustor cup. The cavity around the nozzle ensures that
no atomization energy is dissipated through contact with the adaptor and, if
needed, provides a passageway to cool the nozzle. Unique features of this
nozzle, relative to more conventional atomizers, are low pressure drop (<1 psi),
a limitation on crystal temperature of 300%C, and low droplet momentum. This
nozzle is believed to provide better atomization and mixing with combustion air
than the air-atomized GPU-3 nozzle. Because of the low droplet velocity, care
must be taken in designing how the air enters the combustor.
The fuel nozzle/combustor development work is being conductr.d using MTI's
Free-Piston Combustor Test Rig. In addition to normal instrumentation to deter-
mine flows, temperatures/pressures of incoming fuel and air, combustor
cup/liner sleeve temperature, axial gas temperatures, and gaseous emissions
inside the combustor will be measured. During steady-state testing, engine load
conditions from idle to maximum power were simulated by varying firing rate,
air-to-fuel ratio, and combustor inlet air temperature. Additional testing will
be conducted to determine ianition and blowout limits. Modifications to the
nozzle and combustor will be made to meet the following design goals:
* maximum firing rate of 15 kW;
* adequate ignition and blowout margin;
* 10:1 turndown ratio;
0 nonvisible exhaust plume;
• gas radial temperature profile of: T - T /T - T < .25;
* combustor AP/P < 6%; and,
• combustor efficiency > 99%.
E. CONCLUSIONS OF LIQUID-FUEL EXTERNAL HEAT SYSTEM DESIGN 4
In the liquid-fuel external heat system design and development effort, special
emphasis will be placed on the preheater subcomponent. Since combustion
chemical efficiency is near unity, the preheater is the main determinant of heat
6-32
Adapter
Sono-TekUltrasonic
Combustor Cup
Purge/CoolingAir
Figure 6-23 Proposed EM Diesel Fuel Nozzleand Combustor Cup
822758
6-33
i~ -:
system efficiency, assuming heat lost through conduction to the cold parts of
the engine, and convection losses to the environment, are minimized. Thermal
isolation of the external heat system through the use of insulation is used to
achieve minimal convection losses to the environnant.
The natural-gas-fired EM uses a preheater consisting of a folded fin, 310 stain-
less-steel matrix sandwiched between two ceramic Kaowool cylinders. MTI's ASE
Program experience with burning diesel fuel indicates that stainless-steel
preheaters will corrode if the preheater matrix or average gas temperature in
the matrix is below the sulfuric acid dew point (300*F). Further experience
indicates that no problem exists with unleaded gasoline. For diesel fuel,
increasing amounts of sulfur impose an increasingly corrosive environment. Two
possible solutions to acid corrosion are to prevent acid condensation by raising
the exhaust temperature, or to use an alternate matrix material. The first
approach reduces efficiency; trade-off studies will be performed to determine
the overall system effects.
Engine test experience with the EM Performance and Endurance Engines has led to
refinements of the external heat system's overall design. Proposed changes to
the combustor and fuel nozzle have already been discussed. The design refine-
ments, also applicable to liquid-fuel operation, involved a change of material
for the combustor liner, a different preheater fabrication technique, and detail
design changes to the flame shield, combustor liner, and fin stuffer. (These
pieces are identified in Figure 6-20.)
The combustor liner is currently made of Kaowool, and has a tendency to wear in a
vibratory environment. The relatively soft and porous Koawool will be replaced
with a harder, mechanically stronger ceramic called TabularM (a mixture of
aluminum oxide, silica, and mullite). This material is already used for the
flame shield, where mechanical integrity at high temperature has been demon-
strated. The method of attaching the liner to the external heat system has been
changed to include a more positive, seal-threaded or clamped arrangement at the
bottom, instead of the original method of preloading the liner with the com-
bustor cover. The preheater end rings (U-shaped) will now be brazed directly to
the folded fin matrix to eliminate a potential leakage path. A modification to
the flame shield is the addition of locating ridges to more positively align the
combustor to the heater head, thereby improving the circumferential uniformity
6-34
of heater head temperature profile. The change to a ceramic fin stuffer (used
to enhance heat transfer at the bottom of the heater head) was to strengthen the
locating pip, which has a tendancy to break off.
As mentioned previously, the combustor cup itself will ultimately be made of
ceramic. The feasibility of fabricating the cup out of ceramic TabularTM has
been confirmed through vendor contact. The change to ceramic, however, will not 4be made until aerodynamic developmient of the cup is completed in the Combustor
Rig.
6-35
VII. CONCLUSIONS/RECOMMENDATIONS
'he major objective of this WPAFB program (the design, fabrication, and test of
an FPSE combustor controller) met or exceeded all expectations and goals. The
combustor ccntroller automatically maintains heater head temperatures with
respect to air/fuel ratio and/or engine load clanges, while operating effec-
tively over the entire operating range of the TDE. The same controller system
can be expanded to include the power control and provide for unattended
operation.
The study evaluation of the FPSE for Air Force use (the secondary objective of
the program) indicated that whatever the power control strategy of the
engine/alternator system, it can be integrated with the combustor controller.
In addition, the monolithic heater head design of the FPSE has the potential for
use with military logistic fuels without any anticipated problems. The design
of a liquid-fuel combustor that integrates with an FPSE has been completed, andis presented in this report.
Combustor controller development and FPSE evaluation for unattended operation
provide the base from which FPSE's can be developed for Air Force use. To meet
the particular requirements of the Air Force for its small engine-generator
unattended site operation, the following efforts are recommended:
* development of a liquid-multifuel external combustor system using
identical combustor control strategy developed herein for the
natural-gas-fired system;
* integration of the selected FPSE power control with the combustor
control;
* development of the start-up/shutdown capability of the system
control (power control/combustor control); and,
* demonstration of a stand-alone, controllable FPSE.
7-1 A
A -A
VIII. REFERENCES
1. Dochat, G. R., Moynihan, T. M., Dhar, M.; "l-kW Free-Piston
Stirling Engine/Linear Alternator Test Program," 15th IECEC,
1980.
2. Dochat, G. R., Vitale, N. G., Moynihan, T. M. ; "Development of a
Free-Piston Test-Bed Engine," 16th IECEC, 1981.
3. Lienesch, J. H., and Wade, W. R., "Stirling Engine Progress
Report: Smoke Odor, Noise, and Exhaust Emissions," SAE Paper I680081, January, 1968.
4. Davis, S. R., at. al., "Combustion and Emission Formation in the
Stirling Engine with Exhaust Gas Recirculation," SAE Paper
710824, October, 1971.
'4
8-1 q,0or*" T ~ o o~it i -619.06/98
LlI