...

Daggett2004-H2O-Mist-NOxReduc-Airport.pdf

by user

on
Category: Documents
4

views

Report

Comments

Transcript

Daggett2004-H2O-Mist-NOxReduc-Airport.pdf
NASA/CR—2004-212957
C&EA–BQ130–Y04–002
Water Misting and Injection of Commercial
Aircraft Engines to Reduce Airport NOx
David L. Daggett
Boeing Commercial Airplane Group, Seattle, Washington
March 2004
The NASA STI Program Office . . . in Profile
Since its founding, NASA has been dedicated to
the advancement of aeronautics and space
science. The NASA Scientific and Technical
Information (STI) Program Office plays a key part
in helping NASA maintain this important role.
∑
CONFERENCE PUBLICATION. Collected
papers from scientific and technical
conferences, symposia, seminars, or other
meetings sponsored or cosponsored by
NASA.
The NASA STI Program Office is operated by
Langley Research Center, the Lead Center for
NASA’s scientific and technical information. The
NASA STI Program Office provides access to the
NASA STI Database, the largest collection of
aeronautical and space science STI in the world.
The Program Office is also NASA’s institutional
mechanism for disseminating the results of its
research and development activities. These results
are published by NASA in the NASA STI Report
Series, which includes the following report types:
∑
SPECIAL PUBLICATION. Scientific,
technical, or historical information from
NASA programs, projects, and missions,
often concerned with subjects having
substantial public interest.
∑
TECHNICAL TRANSLATION. Englishlanguage translations of foreign scientific
and technical material pertinent to NASA’s
mission.
∑
∑
∑
TECHNICAL PUBLICATION. Reports of
completed research or a major significant
phase of research that present the results of
NASA programs and include extensive data
or theoretical analysis. Includes compilations
of significant scientific and technical data and
information deemed to be of continuing
reference value. NASA’s counterpart of peerreviewed formal professional papers but
has less stringent limitations on manuscript
length and extent of graphic presentations.
TECHNICAL MEMORANDUM. Scientific
and technical findings that are preliminary or
of specialized interest, e.g., quick release
reports, working papers, and bibliographies
that contain minimal annotation. Does not
contain extensive analysis.
CONTRACTOR REPORT. Scientific and
technical findings by NASA-sponsored
contractors and grantees.
Specialized services that complement the STI
Program Office’s diverse offerings include
creating custom thesauri, building customized
databases, organizing and publishing research
results . . . even providing videos.
For more information about the NASA STI
Program Office, see the following:
∑
Access the NASA STI Program Home Page
at http://www.sti.nasa.gov
∑
E-mail your question via the Internet to
[email protected]
∑
Fax your question to the NASA Access
Help Desk at 301–621–0134
∑
Telephone the NASA Access Help Desk at
301–621–0390
∑
Write to:
NASA Access Help Desk
NASA Center for AeroSpace Information
7121 Standard Drive
Hanover, MD 21076
NASA/CR—2004-212957
C&EA–BQ130–Y04–002
Water Misting and Injection of Commercial
Aircraft Engines to Reduce Airport NOx
David L. Daggett
Boeing Commercial Airplane Group, Seattle, Washington
Prepared under GRC Purchase Order C–74711A
National Aeronautics and
Space Administration
Glenn Research Center
March 2004
Acknowledgments
This document summarizes the efforts of many participants, all of whom were essential to the successful evaluation
of water misting technology as could be applied to future commercial airplanes. The author gratefully
acknowledges the contributions of the Boeing team:
Airport Strategy
Aerodynamics
Configuration
Costs:
Propulsion/maintenance
System
Water costs
Emissions:
Propulsion:
Audit
Performance
Noise:
David Nielson
Gnanulan Canagaratna
Paul Wojciechowski
Al Dunn
Doug Whitfield
Axel Schauenburg
Doug DuBois
Mike Garrison
Matt Naimi
Keith Parsons
William Herkes
Ron Olsen
Andy Ouellette
Weights
In addition, invaluable help from outside Boeing was also provided by:
NASA Glenn Research Center:
Performance
Systems
Engine Companies:
GE Industrial
GE Aero Engine
Pratt & Whitney Engine
Rolls-Royce
CH2M Hill:
Water Purification
Engine Water Injection
Chris Snyder
Jeff Berton
Robert Hendricks
Carl Shook
Richard Streamer
Arthur Becker
Nigel Bond
Bill Farmer
Dan Robinette
David Cheever
Trade names or manufacturers’ names are used in this report for
identification only. This usage does not constitute an official
endorsement, either expressed or implied, by the National
Aeronautics and Space Administration.
Available from
NASA Center for Aerospace Information
7121 Standard Drive
Hanover, MD 21076
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22100
Available electronically at http://gltrs.grc.nasa.gov
EXECUTIVE SUMMARY
This preliminary study shows that by applying a new twist to an old water
injection scheme, engine NOx emissions may be significantly reduced. In addition,
the technology might offer cost savings to airlines as it could have the potential to
save money on engine maintenance.
With every new engine model, pressure ratios have been climbing in the quest
to improve fuel efficiency. New combustors have also been developed to help
offset the exponentially higher NOx that goes with these increased pressure ratios.
The Boeing company, NASA Glenn and the Air Force Research Lab are
working to study and report on how injecting finely atomized (misted) water into the
engine’s low pressure compressor would affect airplane and engine performance.
Water misting evaporates purified water to reduce the temperature of the
engine inlet air and makes for a denser mixture. As opposed to old style water
injection schemes designed for WWII through 1970’s aircraft for thrust
augmentation, and 30-year old commercial water injection, this “water misting”
approach has additional potential benefits of improved SFC, reduced emissions
and greatly reduced turbine inlet temperature. Similar technology is currently used
for industrial gas turbines to increase power output and reduce NOx on hot days.
This task took a preliminary look at system design, airplane performance,
maintenance, and cost implications of using the technology in aircraft for takeoff
and climb-out use only. A specially designed engine performance model, or “deck”,
was used to evaluate the various water injection schemes. Conceptual water
delivery systems were also designed for the airframe.
The study found that applying water misting prior to the LP compressor may
be preferable to older direct combustion water injection systems or where water is
injected into the HP compressor. If the water misting rate could be increased from
a 0.83% water-to-air ratio (present industrial gas turbine rate) to about 2.2%, this
could reduce NOx emissions some 47% from non-water misted engines. On cold
days no SFC penalty would occur. On days above 59F, a fuel efficiency benefit of
about 3.5% would be experienced. Reductions of up to 436 deg R in turbine inlet
temperature were also estimated, which could lead to increased hot section life. A
0.61 db noise reduction was calculated. A nominal airplane weight penalty of less
than 360 lbs. (no water) was estimated for a 305 passenger airplane. The airplane
system cost is initially estimated at $40.92 per takeoff giving an attractive NOx
emissions reduction cost/benefit ratio of about $1,663/ton.
There is a high level of uncertainty to the costs and benefits reported here, but
the results are promising enough to warrant a deeper look at the possibilities this
technology might offer.
NASA/CR—2004-212957
iii
TABLE OF CONTENTS
Page
LIST OF FIGURES ..................................................................................................VI
GLOSSARY.............................................................................................................IX
1.0 INTRODUCTION ................................................................................................ 1
1.1 Study Objective ......................................................................................................... 1
1.2 Work Tasks............................................................................................................... 1
1.3 Potential Benefits ...................................................................................................... 2
2.0 BACKGROUND.................................................................................................. 3
2.1 Environmental pressures........................................................................................... 3
2.2 NOx........................................................................................................................... 3
2.2.1 NOx generation.................................................................................................................... 3
2.2.2 Current NOx reduction methods .......................................................................................... 6
2.2.3 NOx at airport ...................................................................................................................... 8
2.3 Water Injection System Descriptions......................................................................... 9
2.3.1 Traditional engine water injection system .......................................................................... 10
2.3.2 Compressor water misting system .................................................................................... 14
2.3.3 Traditional water injection, airframe system ...................................................................... 16
2.4 Commercialization issues of NOx reduction technologies ........................................18
2.4.1 Maintenance, Reliability and Operability ............................................................................ 18
2.4.2 New Engine/Airplane Introduction...................................................................................... 20
2.4.3 Retrofit ............................................................................................................................... 20
2.4.4 Previous water injection studies -- lessons learned........................................................... 20
3.0 STUDY METHOD ............................................................................................. 22
3.1 Process ....................................................................................................................22
3.2 Airplane and Engine Model ......................................................................................23
3.2.1 Airplane Type..................................................................................................................... 23
3.2.2 Engine Types ..................................................................................................................... 23
4.0 RESULTS ......................................................................................................... 24
4.1
Industrial Engine Performance....................................................................24
4.1.1 Increased Power ................................................................................................................ 24
4.1.2 Same Power, Reduced NOx.............................................................................................. 25
4.2
Aero Engine Performance...........................................................................25
4.2.1 Engine Inlet Injection.......................................................................................................... 26
4.2.2 LP Compressor Injection ................................................................................................... 29
4.2.3 HP Compressor Injection................................................................................................... 30
NASA/CR—2004-212957
iv
4.2.4 Combined LP and HP compressor injection...................................................................... 31
4.2.5 Combustor Injection........................................................................................................... 31
4.2.6 System Comparison ......................................................................................................... 33
4.3 Airframe System Description ....................................................................................33
4.3.1 Airframe system for combustor Injection ........................................................................... 33
4.3.2 Airframe system for LP compressor injection.................................................................... 34
4.4 System Performance Summary................................................................................36
4.4.1 Weights.............................................................................................................................. 36
4.4.2 Thrust................................................................................................................................. 36
4.4.3 Takeoff, Climb and Range Performance ........................................................................... 36
4.4.4 Fuel Use ............................................................................................................................ 38
4.4.5 Emissions .......................................................................................................................... 40
4.4.6 Noise.................................................................................................................................. 47
4.4.7 Maintenance ...................................................................................................................... 48
4.4.8 Water Conditioning and Cost............................................................................................. 48
4.4.9 System Cost ...................................................................................................................... 49
4.4.10 Operating Economics ...................................................................................................... 50
4. 4.11 Airline Operator Survey................................................................................................... 53
5.0 CONCLUSIONS AND RECOMMENDATIONS ................................................ 54
5.1 Summary of Results.................................................................................................54
5.2 Analysis of Results ...................................................................................................55
5.3 Cost Uncertainties....................................................................................................55
5.4 Recommendation .....................................................................................................56
APPENDIX A. NOX CALCULATION METHODOLOGY ....................................... 57
APPENDIX B. COST ESTIMATION OF WATER INJECTION .............................. 59
APPENDIX C. STUDY FEEDBACK ...................................................................... 67
REFERENCES ....................................................................................................... 78
NASA/CR—2004-212957
v
LIST OF FIGURES
Page
Figure 2.1. Combustor inlet temperature increases with compressor pressure
ratio ..................................................................................................... 4
Figure 2.2. NOx increases rapidly as combustor inlet temperature (T3)
increases14 .......................................................................................... 5
Figure 2.3. NOx increases very rapidly for small increases in engine overall
pressure ratio ...................................................................................... 5
Figure 2.4. New GE TAPS combustor technology is reducing NOx formation ........ 6
Figure 2.5. Increasing OPR trends have delineated NOx progress ......................... 7
Figure 2.6. NOx is the airplane emission of focus at airports .................................. 8
Figure 2.7. Landing Take Off (LTO) is used to measure airport emissions ............. 9
Figure 2.8. Higher OPR engines are allowed to emit more NOx ............................. 9
Figure 2.9. First Boeing use of water injection was for early 707 aircraft............... 10
Figure 2.10. Water was injected prior to the combustor via spray bars on early
Boeing 747 aircraft engines............................................................... 11
Figure 2.11. Manifolds supply water to the injection spraybars in older 747
aircraft engines.................................................................................. 12
Figure 2.12. Traditional Industrial-type water injection system illustrated on an
aircraft engine.................................................................................... 13
Figure 2.13. Traditional industrial systems inject water directly into the
combustor ......................................................................................... 13
Figure 2.14. Water misting intercooler system sprays water into LP and/or HP
compressor with HPC air to assist in water atomization.................... 14
Figure 2.15. LPC water injection can move compressor away from surge line27 ... 15
28
Figure 2.16. 747-200 airframe water injection system is well-proven .................. 16
Figure 2.17. 747 water injection system used dry bays in the wings to avoid
displacing any fuel capacity............................................................... 17
Figure 2.18. Installation of water injection tanks in aircraft is a proven technology17
Figure 4.1. Injecting water into LP compressor on an industrial engine during
warm days increases power while reducing NOx, T3 and SFC
(Btu/kW-hr)........................................................................................ 25
Figure 4.2. Evaporating water will reduce air temperature from 100F to 69F and
increase relative humidity from 20% to 100%26 ................................. 26
NASA/CR—2004-212957
vi
Figure 4.3. Evaporating water in inlet increases thrust, reduces NOx and T3 with
little impact on SFC ........................................................................... 27
Figure 4.4. Retarding the throttle to keep constant thrust while using LPC water
injection further reduces SFC, NOx, T3 and T4 ................................ 28
Figure 4.5. Water misting during takeoff either reduces SFC, or increases thrust 28
Figure 4.6. Starting temperature doesn’t make much difference on SFC or T4
reduction as long as injected water can completely evaporate ......... 29
Figure 4.7. Increasing water/air ratio to 2.2% further reduces NOx, T4 and SFC.. 30
Figure 4.8. HPC injection offers less SFC, NOx and T4 benefit than LPC
injection ............................................................................................. 30
Figure 4.9. Combustor water injection requires less water than LPC injection11 ... 31
Figure 4.10. Thermal efficiency and NOx decreases as water injection rate
increases ........................................................................................... 32
Figure 4.11. Injecting water into the combustor increases SFC while decreasing
NOx and T4 ....................................................................................... 33
Figure 4.12. Airframe water system for direct combustion injection system .......... 34
Figure 4.13. Airframe water system for LP compressor injection .......................... 34
Figure 4.14. 150 gallon tank located in each wing dry bay .................................... 35
Figure 4.15. 30% of water (750 lb. for LPC system) is used during takeoff roll, so
climb performance is minimally affected by weight ........................... 37
Figure 4.16. Water misting greatly reduces T4 during the most critical phase of
flight on current technology engines.................................................. 38
Figure 4.17. Combustor water injection uses 51 lb. more fuel from takeoff to
3,560’ altitude while LPC injection uses 90 lb. less fuel than base
engine on standard day..................................................................... 39
Figure 4.18. Water wash is used to clean engine and restore performance......... 39
Figure 4.19. Deterioration can reduce SFC ........................................................... 40
Figure 4.20. ICAO emissions data from the GE90-85 engine (2GE064) was used
to validate the NASA NEPP emissions results50 ............................... 41
Figure 4.21. As little NOx emissions are generated during taxi, water injection
was not used for this phase............................................................... 42
Figure 4.22. NOx is reduced 46.5% during takeoff and climb-out saving 49 lbs. of
NOx emissions to 3,560 feet (beyond normal LTO cycle altitude) .... 43
Figure 4.23. Emissions Relationship...................................................................... 44
Figure 4.24. HC and CO increase with decreasing NOx54 ..................................... 45
16
Figure 4.25. CO generation by water injection is dependant on engine type ...... 46
Figure 4.26. Test results show that smoke may decrease with water injection4 .... 47
NASA/CR—2004-212957
vii
Figure 4.27. Noise decreases slightly because mass averaged jet velocity
decreases.......................................................................................... 48
Figure 4.27. Cost breakdown of airplane water misting intercooler system........... 52
Figure 4.28. Large airplane engine water misting may prove to be substantially
less costly than other industrial NOx reduction technologies. ........... 53
Figure A-1. Using Boeing GE90-85B engine data, the NASA NOx equation under
predicts NOx reduction potential by 7.4% at the 85% thrust level..... 58
Figure B-1: Annual Traffic Seattle Tacoma International Airport 2002 ................... 59
Figure B-2: Sea-Tac Traffic December 2002.......................................................... 60
Figure B-3: Water required per cycle and thrust for various aircraft types.............. 61
Figure B-4. Total Conditioned Water Cost.............................................................. 66
NASA/CR—2004-212957
viii
GLOSSARY
BPR
By Pass Ratio
CAEP Committee on Aviation Environmental Protection (ICAO)
CO
Carbon Monoxide
CFR
Code of Federal Regulations (USA)
DAC
Dual Annular Combustor
EINOx Emissions index for NOx given as grams of NOx/Kg fuel
FAR
Federal Aviation Regulation (USA)
GE
General Electric
HC
Hydro-Carbons
HP
High Pressure
HPC
High Pressure Compressor
ICAO International Civil Aviation Organization
kg
kilogram
kts
nautical miles per hour
lb
pound
Load Factor Percentage of an airplane's seat capacity occupied by passengers
LTO
Landing Take-Off cycle
LP
Low Pressure
LPC
Low Pressure Compressor
MTOW Maximum Take-Off Weight
NASA National Aeronautics and Space Administration (USA)
NOx
Nitrogen Oxides
NMI
Nautical mile
OPR
Overall Pressure Ratio
OEW Operating Empty Weight
P&W Pratt & Whitney
PAX
passengers
SLS
Sea Level Static
std
Standard
SFC
Specific Fuel Consumption (lb. fuel per hour/ lb. thrust or power)
T3
Temperature at the exit of the HP compressor
T4
Temperature at the inlet to the high pressure turbine (TIT)
TIT
Turbine Inlet Temperature
TOGW Take Off Gross Weight
TSFC Thrust Specific Fuel Consumption
NASA/CR—2004-212957
ix
1.0 INTRODUCTION
This report documents the results of a NASA-funded research study to
evaluate the airplane performance impacts of water injection technology.
1.1 Study Objective
Can new industrial gas turbine water injection schemes, used for NOx
reduction, be used on future aircraft for cost and performance improvements?
Emissions are playing an increasingly important role in the design of
commercial aircraft as well as transport military aircraft. It is important to evaluate
water injection technology because absolute NOx emissions from aircraft have
been difficult to control. In some cases airport NOx emissions have increased even
with the introduction of newer aircraft.
In this study, preliminary costs to operators of using water injection technology
for airport NOx reduction will be weighed against the cost of ever increasing
1
emissions-based landing fees. Additionally, a side benefit of water injection is to
reduce engine turbine inlet temperatures. This might extend engine hot section life
that could conceivably offset any costs incurred from operating the water injection
system. Other maintenance issues also will be addressed in an attempt to assess
the cost of the entire system.
1.2 Work Tasks
To start the study, a search was done of public and Boeing internal
documents on past water injection work. Extensive writings have been published
on the emissions reduction potential of this technology.2,3,4,5,6,7
NASA Glenn research center conducted combustor emissions tests jointly
with the Air Force Research Laboratory to establish emissions reduction potential of
water misting technology. This was used in the study to estimate NOx reduction
potential as well as preview the potential for HC and CO emissions increase.
NASA Glenn research center modified a NASA Engine Performance Program
(NEPP) to gather overall performance estimates for injecting water before the LP
compressor, between the LP and HP compressors and directly into the combustor.7
These performance models were verified with several performance points supplied
by models from the GE Power Systems group (industrial engine) and the propulsion
group at Boeing Commercial Airplanes.
The Boeing Product Development Group performed a conceptual design of
the airframe water delivery system. This included design layout, weight and cost
estimates for a future technology new production airplane.
Airplane performance estimates of a future technology 305 passenger
airplane equipped with water injection were conducted by the Boeing aerodynamics
group.
NASA/CR—2004-212957
1
The environmental and emissions groups at Boeing provided emissions and
water use estimates for the aircraft mission as well as estimated water cost, airport
servicing costs and gathered airline operator feedback for using such a system.
GE, Pratt & Whitney and Rolls-Royce engine companies were all solicited to
provide input and feedback on a draft report.
1.3 Potential Benefits
This technology fits with NASA’s vision of improving the quality of life here on
earth. Some of the benefits that may be enjoyed by the aviation community when
using this technology are to:
-
provide a cost effective airport NOx emissions control technology that may
allow continued growth of aviation.
-
may allow combustors to be optimized for cruise NOx reduction instead of
compromising on a balance of cruise and takeoff reduction.
-
possibly increase engine hot section life and reduce overhaul cost.
-
may promote compressor cleaning to prolong engine performance which
could help to reduce fuel use.
-
reduce potential fuel use penalties and associated risks of other NOx
control technologies, such as direct combustor water injection, or Dual
Annular Combustors.
These potential benefits will be explored in the following study and weighed
against the liabilities and uncertainties of the system. A high-level estimate will be
given of the total system performance.
NASA/CR—2004-212957
2
2.0 BACKGROUND
Old style water injection systems used on early Boeing 707 and 747 aircraft
for thrust augmentation were unpopular with airlines because little benefit was
readily seen while the drawbacks of servicing the system with water were observed
every day. However, as the current drawbacks of emissions landing fees and
airport restrictions overpower the need for servicing a water injection system, water
injection could once again become popular.
2.1 Environmental pressures
The emissions of regulatory attention tend to be gaseous engine pollutants
(e.g. NOx, CO and HC) and increasingly, microscopic smoke particles and carbon
dioxide (CO2). International regulations make allowances for more fuel efficient,
higher pressure ratio engines to emit more NOx emissions than older engines. This
is because higher NOx emissions are typically traded off for lower HC, CO and CO2
emissions. At the same time, airports are increasingly faced with increasing
pressure to control NOx emissions from all sources, and in some cases are facing
caps from local regulatory agencies and these may start to limit some airline
operations.8
Aviation related emissions of nitrogen oxides, which contribute to the
formation of ozone, have been of particular concern to many airport operators. A
federal study at 19 airports estimated that by 2010, aircraft emissions have the
potential to significantly contribute to air pollution in the areas around these
airports.9 In response to a doubling of aircraft operations from 1976 to 2000,
European airports and several US local regulatory agencies are implementing
emissions-based landing fees and airport emissions caps that are limiting traffic
growth.10 The US military are also faced with pressures to reduce aircraft
emissions and have expended emissions R&D funds equivalent to other
11
government organizations. Lastly, improved knowledge of health effects of
emissions has led to increasing valuations in practically all emissions.12
2.2 NOx
2.2.1 NOx generation
The generation of NOx gasses are closely linked to the engine combustor
flame temperature that is in turn influenced by the Overall Pressure Ratio (OPR) of
the engine’s compressor. However, engines that have high pressure ratios are
desirable since this tends to reduce Specific Fuel Consumption (SFC). Thus, SFC
gains are often traded off against increased NOx emissions.
During compression of air from the inlet of the engine to the inlet of the
combustor, a temperature rise occurs as work is imparted to the fluid (air). The less
efficient the compressor, for example 80% versus the ideal of 100%, the higher the
ending temperature as shown in Figure 2.1.13
NASA/CR—2004-212957
3
2000
80% Efficiency
1800
1600
Temperature (R)
1400
100% Efficiency
1200
1000
800
600
400
Future
Engines
Present
Engines
200
0
1
10
20
30
40
50
60
Pressure Ratio
Figure 2.1. Combustor inlet temperature increases with compressor pressure ratio
After compression of the air by the compressor and introduction into the
combustor, high temperatures oxidize the nitrogen in the air into oxides of nitrogen,
collectively called “NOx.” This process occurs at temperatures above 1800K flame
temperature and progresses rapidly as the temperature increases (film cooling on
the combustor wall prevents the metal structure from melting). Combustor flame
temperature generally increases with increased combustor inlet temperatures.
Figure 2.2 shows the relationship of combustor inlet temperature (and hence flame
14
temperature) to NOx formation.
For this study, a general emissions equation was used to predict how much
NOx would be generated, based on the combustor T3 and P3 conditions. This
NASA equation is listed below as equation 1. A further analysis and validation of
the equation, along with other equations are listed in Appendix A.15
EINOx = 33.2*((P3/432.7)^0.4)*EXP((T3-459.67-1027.6)/349.9+(6.29-6.30)/53.2)
(Equation 1)
where:
P3 = Pressure of compressor exit (psia)
T3 = Temperature of compressor exit (Deg R)
When taking into account the rise in temperature with engine pressure ratio,
and the rapid rise in NOx with combustor inlet temperature, a very rapid rise in NOx
occurs with small increases in pressure ratio. To see this relationship, a modern
large engine (e.g. GE90 or PW4000 type of engine) performance deck was
manipulated to increase OPR and observe the impact on SFC. Using the above
equation to predict NOx impact, one can see in Figure 2.3 that increasing OPR
results in small improvements to SFC but results in large increases in NOx.
NASA/CR—2004-212957
4
Figure 2.2. NOx increases rapidly as combustor inlet temperature (T3) increases
30
ICAO NOx Parameter
20
10
0
DD99-33.xls
Percent Change from Baseline
40
TSFC
-10
0
15
30
Pressure Ratio Increase (%)
Figure 2.3. NOx increases very rapidly for small increases in engine overall
pressure ratio
NASA/CR—2004-212957
5
14
2.2.2 Current NOx reduction methods
Efforts to reduce NOx emissions have resulted in research and development
programs to introduce low NOx combustor technology to aero gas turbine engines.
The design philosophy behind these combustors is to quickly blend the atomized
fuel and air mixtures very well prior to its being burned in the combustor. As we
saw in the previous section, high flame temperatures inside the combustor result in
high NOx emissions. Older combustors, although efficient, stable, reliable, and
often low in HC and CO emissions, had fuel/air pockets within the combustor where
very hot gas generated large amounts of NOx. The combustion products then
needed to be cooled via air dilution holes just prior to its leaving the combustor in
order prevent the melting of the nozzle guide vanes and high pressure turbine
blades. Newer combustors, such as the one shown in Figure 2.4, mix the air and
fuel very well in the dome of the combustor to achieve a more homogenous
mixture, thereby eliminating the hot pockets within the combustor. Since the flame
temperature is more accurately controlled, and overall is cooler, the dilution holes
are eliminated. The introduction of these low NOx combustors is vital to help
control emissions over the entire range of the aircraft.16
Figure 2.4. New GE TAPS combustor technology is reducing NOx formation
NASA/CR—2004-212957
6
Although the new combustor technology is capable of reducing NOx
emissions, it generally arrives just in time to be introduced into a new engine with
an even higher OPR. Often the new combustor only offsets the additional NOx that
would have been generated by the higher OPR. Continued development of
advanced combustor concepts are needed for cruise NOx emissions reduction and
renewed investigation of other concepts such as water injection for takeoff
emissions reduction.
Since high fuel efficiency turbine engines are very desirable, the focus in the
aviation community has been on increasing OPR for newer engines. Without the
use of improved low emissions combustors, NOx emissions would have climbed
exponentially. However, by introducing these new technology combustors into the
new, higher OPR engines, NOx emissions have been kept in check. Figure 2.5
shows these OPR and NOx performance trends for small commercial aircraft over
time. This story is similar for other aircraft categories. As a result, little progress
has been achieved in reducing airport NOx emissions because the focus has been
on reducing fuel use (i.e. CO2) emissions and operator cost.
35
0.1
30
0.07
25
Average Pressure Ratio Trend
0.06
20
0.05
Average LTO NOx Trend
0.04
737-500
727-100
DC9C-30F
0.03
737-600
15
717
737-200
737-300
10
Engine Pressure Ratio
0.08
0.02
5
0.01
0
1960
1965
1970
1975
1980
1985
1990
1995
2000
2005
Small Commercial Airplane Certification Date (year)
Figure 2.5. Increasing OPR trends have delineated NOx progress
NASA/CR—2004-212957
7
0
2010
DD98-10.xls
NOx emissions (gm/LTO/pax)
0.09
2.2.3 NOx at airport
Although effort is under way to establish NOx emissions regulations for cruise,
all current regulations are intended to constrain emissions in the airport vicinity.
Airplanes are typically the largest contributors to airport NOx. A
representative comparison of emissions type and mode is shown in Figure 2.6.17
By halving airplane NOx emissions, other modes of transportation (i.e. car) could
conceivably overtake airplanes as the major emissions contributor.
The standard method of calculating airport emissions for aircraft is the
Landing Take Off (LTO) cycle18 as shown in Figure 2.7. Established times in
modes are set for each operating condition (idle, taxi, takeoff and approach). The
fuel flow (kg. /min.) and emissions index (grams of emission per kg of fuel
consumed) at each operating condition is measured during the certification process
of the engine model. Summing up these values and dividing by the engine’s Sea
Level Static (SLS) takeoff thrust produces a result titled “Dp/Foo”. This is used in
evaluating the emissions vs the regulatory standards.
Airport Emissions
for 5 sq. mi.*
80
70
Sample US
Airport
Emissions Contribution (%)
60
50
40
NOx
30
HC
20
10
CO
0
Airplane
* Sample 1992 airport using EDMS
GSE
Car
Parking
Source
Figure 2.6. NOx is the airplane emission of focus at airports
NASA/CR—2004-212957
8
Power setting
% of std. day take-off (T/O) thrust
Idle: 7%
Take-off: 100%
Climb: 85%
Approach: 30%
Climb-out
3000 ft
Take-off
Taxi-in
Approach
Taxi-out
Operating mode
1. Taxi / idle
2. Take-off
3. Climb
4. Approach
Time
26.0 minutes
0.7 minutes
2.2 minutes
4.0 minutes
•
•
•
•
Fuel Flow
x.x kg/min
x.x kg/min
x.x kg/min
x.x kg/min
EI
x.x g/kg fuel
x.x g/kg fuel
x.x g/kg fuel
x.x g/kg fuel
•
•
•
•
=
=
=
=
Dp/Foo
x.x g/kN
x.x g/kN
x.x g/kN
x.x g/kN
Sum ÷ Τ/Ο Thrust = Ave. Dp/Foo
80
75
Not Allowed
70
65
mit
4 Li
P
E
CA
60
Allowed
55
50
45
40
20
22
24
26
28
30
32
34
DD99-43.xls
Landing Takeoff Cycle NOx -- Dp/Foo (g/kN)
Airport NOx Emissions
Figure 2.7. Landing Take Off (LTO) is used to measure airport emissions
Overall Pressure Ratio (OPR)
Figure 2.8. Higher OPR engines are allowed to emit more NOx
As will be discussed in the next section, NOx emissions tend to increase
dramatically with increases in the engine’s OPR. Recognizing this relationship, and
taking into account that higher OPR engines typically exhibit better fuel efficiency,
the ICAO regulatory agency has made allowances for high OPR engines to emit
higher NOx emissions. This is shown in Figure 2.8.
2.3 Water Injection System Descriptions
Water injection has been used for over 30 years in industrial engines to
reduce NOx emissions. It has also been used for over 45 year in Boeing’s 707 and
NASA/CR—2004-212957
9
747 aircraft engines to augment thrust some 10-30%.19 However, water injection
has not been used on aircraft to reduce emissions. As gas turbine engines have
matured and became capable of generating ever more thrust, water injection for
new engines has been abandoned. However, there are still a few aircraft in service
that continue to use water injection.20
2.3.1 Traditional engine water injection system
2.3.1.1 Original Pratt & Whitney aircraft systems – At Boeing, water injection
was first used over 45 years ago on Pratt & Whitney JT3C-6 engines. These
engines were installed on early Boeing 707-120 Stratoliner aircraft and the water
injection system augmented takeoff thrust (Figure 2.9). As the water was injected
into the engine inlet, it cooled the air by evaporation and provided a 35% thrust
increase on a 90F day. On days below 40F, water was injected into the HPC
diffuser only which still provided a slight increase in thrust. However, as the
ambient temperature dropped below 22F, no thrust increase could be achieved
from water injection.
This system used a belly tank to store the demineralized water and an
electrically driven boost pump delivered water to the 4 engines. At that point, an
engine-driven mechanical pump then increased the pressure to about 400 PSI for
injection. The engine pumps were know for generating pressure surges and
oscillations which were later corrected through several service bulletins.
Figure 2.9. First Boeing use of water injection was for early 707 aircraft.
NASA/CR—2004-212957
10
The last water injection system to be used on Boeing aircraft was for the early
747-100 and 747-200 aircraft using Pratt & Whitney JT9D–3AW and –7AW series
engines. In this application, water was injected into the compressor discharge air
stream via spray bars located just up steam of the combustor and downstream of
the HP compressor as shown in Figure 2.10.
Water Tank
Figure 2.10. Water was injected prior to the combustor via spray bars on early
Boeing 747 aircraft engines.
NASA/CR—2004-212957
11
Water was supplied to these spray bars from a water manifold that was run
next to the fuel manifolds and is shown in the figure below.
The design of this system suggested that the water distribution was not as
well controlled as in later industrial water injection systems. This would lead to
some portions of the combustor receiving more water which would lead to poor
temperature pattern factors for the HP turbine. Thermal stressing of the case and
surrounding metal structures was also reported on such systems, presumably due
to the sudden introduction of the cool water which then impinged on the hot metal
surfaces.
Figure 2.11. Manifolds supply water to the injection spraybars in
older 747 aircraft engines.
NASA/CR—2004-212957
12
2.3.1.2 Common industrial combustor injection systems – On later industrial
engines, the improved water injection technique was to spray water directly into the
combustor dome (Figure 2.12) via a dual fuel/water nozzle as shown in Figure 2.13.
By atomizing the fuel and water together inside the combustor, a better
distribution of water could be maintained as compared to the previous system
(2.3.1.1), and so the combustor exit thermal pattern factor was restored to
acceptable levels. It also eliminated the thermal stressing on the case since water
was now only directed to where it was needed … inside the combustor.
Water Tank
Figure 2.12. Traditional Industrial-type water injection system illustrated on an
aircraft engine.
Water
Injection
Port
Fuel
Port
Fuel
Water
Fuel
Figure 2.13. Traditional industrial systems inject water directly into the combustor
NASA/CR—2004-212957
13
Other newer methods of injecting water involve the injection of steam through
a special dual port fuel nozzle. However, the injection of water is preferable over
steam as it is more effective in reducing NOx emissions. Since steam injection
requires a larger volume, an especially good optimization of the flow conditions are
required.21 Due to its size and steam generating requirements, steam injection is
not an option for airborne applications.
2.3.2 Compressor water misting system
2.3.2.1 Description – The maximum power that an engine develops is largely
determined by volume (i.e. mass) and the incremental velocity (i.e. acceleration) of
the airflow moving through the engine.22 When water is sprayed into the
compressor inlet, the temperature of the compressor inlet air is reduced and
consequently the air density and thrust are increased.23,24
With the evaporation of the water droplets, and the corresponding drop in air
temperature, the combustor inlet temperature also drops. This reduces NOx
formation. In addition, as the engine thrust has now been increased by increasing
the mass flow, the engine throttle can be reduced to keep the same level of thrust
as before water misting. This decrease in throttle setting also lowers the combustor
inlet temperature. This results in a further drop in NOx formation.
Figure 2.14 shows the water misting system with an injection point before the LP
compressor and also before the HP compressor. Typically, 24 air-assisted spray
nozzles inject the water from the front frame of the engine.46 In addition, water can
also be injected between the LP and HP compressors. The LP compressor
injection system would no doubt be discontinued during very cold atmospheric
conditions to prevent the water from freezing. In the Figure below, high pressure air
from the HP compressor exit could be used to further atomize the water injection
points.
Water
Figure 2.14. Water misting intercooler system sprays water into LP and/or HP
compressor with HPC air to assist in water atomization
NASA/CR—2004-212957
14
2.3.2.2 Operability Concerns – Depending on the location that water is
introduced into the engine, the low pressure compressor and high pressure
compressor can have different operating impacts, either moving towards or away
from the compressor surge line. Figure 2.1525 shows that for both LP and HP
compressor, injection of water into the HP compressor diffusor, (up stream of the
combustor) will result in the compressor moving towards the surge line. However,
when introducing water into the inlet of the LP compressor, this will cause the HP
compressor to move towards the surge line and the LP compressor to move away
from the surge line.
Figure 2.15. LPC water injection can move compressor away from surge line27
NASA/CR—2004-212957
15
The amount of water, the state of the water (evaporated or liquid phase) and
the ambient air conditions will all have an impact on whether the LP compressor
moves towards or away from the surge line. A more in-depth analysis is required to
evaluate how the LP and HP compressors will behave with the amount of water
required to achieve the NOx reduction goal. Once the impact is understood, the
engine could be designed to operate with these increased or decreased surge
margins. The re-designed compressor’s performance and weight impact could then
be taken into account for an overall airplane-level performance assessment.
2.3.3 Traditional water injection, airframe system
The original system used on the 747-100 and -200 airplanes used four
electrically driven (400 Hz, 115/200 VAC, 3ϕ, 36 KVA) high pressure (534-750
psig), high volume (26K-30K pph) centrifugal pumps to inject water directly into the
diffusor section (upstream of the combustion chambers) of the JT9D engines. The
pumps are mounted to water storage tanks fitted with bladders in the wing center
section forward dry bay of 747-100 series airplane, or in a storage tank in the
inboard leading edge of a 747-200 airplane as shown in Figures 2.16,26 2.17 and
2.18. All tanks are equipped with fill, drain and quantity indicating systems.
Remote Service
Panel
Quantity
Indication
Manual Tank
Select Fill
Valve
Fill Connection
Drain Valve
Light
Drain Valve (2)
Integral Water Tank
300 Gallons Water (2)
High Pressure
Pump (4)
t
Line to
Engine No. 2
v
p
p
Manifold
Drain
Valves (2)
Line to
Engine No. 1
Tank
Fill Lines
v
p
p
Flow
Regulator
t
16
Anti
Siphon
Valve
(4)
Front
Spar
(Ref.)
Figure 2.16. 747-200 airframe water injection system is well-proven
NASA/CR—2004-212957
Shutoff
Valve
Line to
Engine
No. 3
Drain
Mast
Flow Pressure
Switch (4)
NO 3
Engine
Tank Quantity
Transducer (2)
Line to
Engine No. 4
28
Figure 2.17. 747 water injection system used dry bays in the wings to avoid
displacing any fuel capacity.
Figure 2.18. Installation of water injection tanks in aircraft is a proven technology
NASA/CR—2004-212957
17
A water injection switch in the cockpit is turned on just before takeoff to
activate the water pumps. As the engine throttles are advanced past 92°,
compressor discharge air will energize the shut off valves and water will flow to the
engines.
When the water is exhausted (about 2 ½ minutes later), a low water pressure
switch will notify the flight crew and also send a signal to the engine fuel control unit
to reduce the fuel flow (and thrust) to normal dry rate to avoid burning up the
turbine section. The flight engineer should then turn the water injection switch off
and turn the “drain valve” switch on. This will drain any remaining water in the
tanks, and lines (about 20 gallons) overboard through a heated drain mast in
approximately 8.5 minutes.
For refill operation, the tanks are connected to a 30 psig external water line
where it will take 5.6 minutes to refill the tanks with 600 gallons of purified water.
The system could operate at temperatures down to 0ºF.
2.4 Commercialization issues of NOx reduction technologies
2.4.1 Maintenance, Reliability and Operability
Maintenance of low emissions systems must also be included in evaluating the
cost of such systems as this may detract substantially from the cost effectiveness of
the emissions control device, and in some cases turn an apparently attractive
technology into an unpalatable one.27
Several water injection airframe system anomalies were found on the earlier
747-100 series aircraft that were later corrected through service bulletins. These
included shutoff valve, water flow regulator, reset check valve, anti-siphon valve,
drain mast reactivation, and installation of 20 micron water filters.28 Other reported
problems were related to leaking fittings.
Another design issue for the airframe system is the requirement to design such
a system as to prevent the mixing of water and fuel in the separate tanks. In 1973,
a BAC 111 aircraft crashed after takeoff at Hamburg Germany because fuel had
29
inadvertently been put into the water tank. Design of unique filling nozzles should
help solve this problem.
For aero engines, turbine blade erosion was identified on early Pratt & Whitney
JT9D-3AW engines, but was corrected on later engines by introduction of highstrength, nickel alloy turbine blades.29 Other hot section problems occurred by not
using de-mineralized water. This has ruined engines which would then require a
complete overhaul.30
On early industrial engines while using water injection continuously, shortened
hot section life was reported for engines using the direct combustor injection
systems.54 In some cases, combustor life was reduced from the typical 16,00024,000 hours to as short as 3,000-8,000 hours. Shortened fuel nozzle life was also
reported. As the components failed, this lead to the failure of the turbine blades,
requiring the engines to be overhauled at substantial cost. As these issues have
NASA/CR—2004-212957
18
now been addressed, engine reliability has increased and no adverse impacts on
engine hot section life or durability have been uncovered.51
On later industrial engines, using the GE Sprint system of continuously injecting
water into the LP compressor for power augmentation, metal erosion has been
discovered on the first three rows of the compressor blades.31 However, for use in
aero engines during the short time span of takeoff, this should not cause a problem.
With the old water injection system on 747 aircraft with JT9D engines where
water was injected in the HP compressor diffusor prior to the combustor (see
section 2.3.2.1), P&W states that there were water leaks issues, case distortion,
combustor and turbine erosion, performance deterioration from hole plugging,
coating erosion and tip clearance problems with the old system. Pattern factor was
also affected. Lastly, water injection into the combustor was preferred over LP
32
compressor injection due to compressor erosion problems. These aspects need
to be examined further for the water misting intercooler approach.
The compressor blade erosion problem, both reported by GE and P&W, when
injecting water into the LP compressor needs to be further investigated. If the
compressor blades are experiencing erosion when water injection is used
continuously, then perhaps there may be some opportunity for cleaning of the
blades when water injection is used only intermittently during takeoff.
On Naval gas turbine engines that had water injection with large water
manifolds, engine flameouts were reported during rapid emergency deceleration
from full power to idle conditions.33 However, this problem was reported to be
solvable by altering the engine control laws but would likely need much further
investigation to examine the much more critical operating envelope for aero engine
applications.
From Boeing maintenance data of TWA aircraft using water injection on 747
aircraft, and from previous 707 aircraft water injection maintenance data, reported
maintenance for the water injection system was $6,865 in 1975 dollars as shown in
table 2.1 below. Calculated current costs are updated by using the consumer price
index.34
Table 2.1. Historical and Current Maintenance Costs
Airplane
Maint. Cost
# takeoffs
Cost per takeoff
Reported 747
$6,865
828
$8.29 (1975 dollars)
$20.62 (2003 dollars)
NASA/CR—2004-212957
19
Water freezing in the lower part of the water tanks and pumps was
reported on early 747-100 aircraft. No permanent damage was reported and the
faulty valve or switch that prevented water from being used or drained was replaced
and the system put back into service.
On freezing days, it was reported that water should not be loaded onto the
airplane prior to 1 hour before departure.28 Presumably, the relatively high
temperature of the water kept the system from freezing. On older aircraft (e.g.
WWII airplanes), water was often mixed with alcohol to prevent freezing on cold
days. This could also be an option for new commercial aircraft that do not use
engine bleed air for passenger cabin pressurization as no alcohol or water vapor
would have a path into the passenger cabin. The alcohol would only be consumed
in the combustor so there would be no additional cabin or airport environmental
issues to consider.
2.4.2 New Engine/Airplane Introduction
For this study, we consider the introduction of the technology to be on newly
designed future commercial airplanes and engines. This traditionally tends to be
the most economical way to introduce new technologies. In addition, when
introducing new technology, the aircraft can be designed to take advantage of any
performance opportunities. This will lead to a further enhancement of the
technology by designing the airframe specifically to the task, multiplying the benefit
which can lead up to a further 50% improvement over the original technology
17
improvement.
For military applications, large cargo aircraft might consider the technology,
but it would be impractical for combat/tactical aircraft.35
2.4.3 Retrofit
Past studies have shown that retrofitting existing aircraft tends to be much
40
more expensive than the original technology designed into new airplanes. As
water injection technology is a very integral system in the engine, it is cost
prohibitive to remove existing aircraft engines and modify them for water injection.
Today, many engines do not routinely undergo complete overhaul at which time
might provide an opportunity to replace existing components with those designed
for water injection. Instead, the engines health and component integrity are
monitored and those modules replaced when needed. Thus, some engines can
stay on wing until the end of the airplane’s life.36
2.4.4 Previous water injection studies -- lessons learned
In 1973, the U.S. EPA promulgated strict emissions control requirements for
37
aircraft engines. As low emissions combustor development was still in its infancy,
alternate means for emissions reduction were sought after. Thus, a cost/benefit
study was conducted to evaluate conventional airplane combustor water injection
systems.38 Often, choosing challenging ground rules of a study can adversely
affect the study results. In this case, the study chose to include water injection for
NASA/CR—2004-212957
20
the APU.39 This was a noble effort as APUs can also contribute measurable
amounts of NOx in the airport environment.40 However, this design ended up
severely penalizing the overall airplane performance. For future studies, the
lessons learned from this earlier endeavor would be to:
1) Eliminate the APU water injection system, reducing total aircraft system
weight some 25%
2) Do not carry water to the destination for use in water injection during
descent, taxi-in and gate arrival, saving some 750 lb on a 747-sized aircraft.
3) Evaluate keeping the engine’s water-to-fuel injection ratio at or below a
0.5:1 ratio to prevent large increases in HC and CO. This also reduces the NOx
reduction effectiveness somewhat, but overall system performance will most likely
improve.
4) Utilize the dry bays in the aircraft wings (see section 2.3.4) for water
storage to avoid having to construct special water tanks in the cargo area, saving
some 50% on the remaining system weight.
5) Utilize improved engine water injection schemes (such as water misting
injection) to avoid the large SFC penalties estimated for the previous study. The
SFC penalty is typically 3% for modern combustor water injection systems and will
probably reduce substantially for the water misting intercooler system. The older
Pratt & Whitney style water injection system in the study was estimated to
contribute a 10% SFC penalty.
NASA/CR—2004-212957
21
3.0 STUDY METHOD
3.1 Process
Historical combustor and engine test data was gathered and compared to
more recent tests41 of advanced combustors. These data were used to first
establish a correlation between the water injection rate and NOx reduction rate.42
Ratios of water to air and water to fuel were calculated and used to predict the
amount of NOx reduction possible for a modern commercial aircraft engine.
Engine performance decks from Boeing (EDASA), GE power systems and
NASA Glenn (NEPP) were used to estimate the performance impact of injecting
water into the engine. These decks were each run with similar fuel, air, water and
power rates to validate that they were providing similar answers. The NASA deck
was then run with water injection rates higher than was possible for the Boeing and
GE decks, setting out to achieve a 50% NOx reduction goal. The NOx reduction
amount was compared to the historical data and found to agree fairly close.
For the now established water injection rate, the airframe systems and water
tanks were designed to inject enough water for a 777-sized aircraft to takeoff and
reach 3,000 feet altitude before exhausting the water supply. The increased
available thrust was not used in takeoff since the aircraft should be designed for fail
safe operation of the water misting system (in the event of a single engine failure or
other critical episode where additional thrust is needed, the water misting or
injection system could enhance safety margin.) Weights, costs and airplane
performance data were then generated.
The change in airplane performance was estimated from the engine deck data
(e.g. SFC change during takeoff) and the airframe design changes (e.g. increased
weight of the system causing higher fuel use.) Airplane performance sensitivities
were used to calculate the change in mission length and fuel use.
Water costs were estimated from historical data as well as input obtained from
water conditioning companies. Airport infrastructure issues were estimated from
internal data and consultation with a major airport operator (Seattle-Tacoma
International).
Customer input was gathered from questionnaires sent to major air carriers to
assess the desirability of the water misting intercooler system.
Particular study emphasis was placed on the water misting intercooler system
as previous engine company studies highlighted many negative aspects of
conventional water injection systems (e.g. SFC penalty, pattern factor).31, 32
NASA/CR—2004-212957
22
3.2 Airplane and Engine Model
3.2.1 Airplane Type
The airplane used for the study was a conceptual 777-200ER aircraft with a
new composite wing sized to be used with a current technology GE90 series engine
as shown in Figure 3.1. This aircraft was previously configured for a NASA Langley
study of 21st century wing technology43 and had a Gross Take Off Weight of about
636,500 lb.
777 Fuselage, Empennage
Generic Current
Technology 85K lb.
Thrust Powerplants
Updated LE, TE
Composite Wing Box,
Alum. Rib Chords,
High AR wing
8% Span Raked Wing Tips
Figure 3.1. Advanced technology 777-type study airplane used for baseline
3.2.2 Engine Types
Two engine types were used in the study, an aero-derivative industrial gas
turbine engine and an aero engine.
3.2.2.1 Industrial Engine Model – The GE aero derivative engine, model LM6000
was used in the study to compare performance with aero engines. This is a
40MW class engine that operates both with and without the Sprint water misting
intercooler system where water is atomized and sprayed into the compressor.
3.2.2.2 Aero Engine Model – A generic, current technology, large bypass ratio
Numerical Engine Performance Program (NEPP), similar to the PW4000 and
GE90 series engines, was used by NASA Glenn and Boeing in the final
performance analysis of the aero engines so that no proprietary engine company
data would be disclosed. GE90-85B and PW4084 engine performance models
were used internally by Boeing to validate the results of the NASA performance
analysis.
NASA/CR—2004-212957
23
4.0 RESULTS
The impact of water injection on engine performance is evaluated for an
industrial engine using a performance program where atomized water is injected
into the LP compressor. In addition, water injection impact was also evaluated for
an aero engine where water is injected directly into the combustor, before the
engine inlet, into the LP compressor, into the HP compressor and a combination of
LP and HP compressor water injection.
A preliminary airframe system was designed and airplane performance was
estimated using aerodynamic performance modeling tools.
4.1 Industrial Engine Performance
For a GE power systems LM6000 industrial engine, the following data was
estimated using their engine performance models. This Sprint water misting
intercooler system injects atomized water before the LP compressor.
4.1.1 Increased Power
The water misting system used on the industrial engine is primarily intended to
boost output power on hot days. It does this by lowering the turbine inlet
temperature (T4), which allows increased fuel flow, bringing the power back up to
cool-day conditions. Thus, a constant T4 temperature is maintained as ambient
temperature increases. As shown in Figure 4.1., at a temperature of 90F, a 20%
increase in power is achieved when water is injected at a rate of 0.87% water to air
mass flow ratio into the engine core. This drops to a 5% increase at 59F for an
injection rate of 0.53%. Below 45F, power increases are negligible.44 Even when
power is increased, the SFC (Btu/kW-hr, LHV), NOx emissions and compressor exit
temperatures (T3) all decrease.
NASA/CR—2004-212957
24
0
25%
59F Day
0.53% water misted
-5
20%
90F Day
0.87% water misted
-10
10%
-15
5%
-20
T3 Decrease (Delta R)
15%
-25
0%
Power
-30
SFC
EINOx
dld03-17.xls
-5%
T3
-35
-10%
Figure 4.1. Injecting water into LP compressor on an industrial engine during warm
days increases power while reducing NOx, T3 and SFC (Btu/kW-hr)
4.1.2 Same Power, Reduced NOx
The above condition in 4.1.1 assumed that power was increased by increasing
water flow rate and T4 was maintained. When maintaining a constant power
setting, and letting T4 fluctuate, it is anticipated that the T3, EINOx and SFC will
further improve. However, as the LPC water misting system on the industrial
engine was designed only to improve power output, the engine performance
models were unable to run this condition.
The same power, reduced NOx scenario is the same as will be considered for
the following aero engine evaluation. Namely, engine power will not be increased
beyond the normal rated engine output, but water injection will instead be used to
reduce NOx and the fallout effects of SFC and T4 will be observed.
4.2 Aero Engine Performance
The aero engine performance effects will be evaluated by varying the method
of water injection (e.g. inlet, LPC, HPC and combustor). The inlet and LPC
injection methods should reflect similar trends as the industrial engine.
NASA/CR—2004-212957
25
4.2.1 Engine Inlet Injection
The first water injection scheme will involve injection of water at the inlet of the
engine. This method is not considered to be feasible for current aircraft engines
due to the significant amounts of water required (largest part of water would exit
through fan and not affect engine core). However, using the Boeing engine
performance deck to validate the industrial engine performance trends is of interest.
Air input conditions to the Boeing engine performance deck were manipulated
to simulate water misting, with complete evaporation, into the inlet of the engine.
This was done by specifying air temperature and humidity conditions. For example,
the psychrometric chart in Figure 4.2 shows two conditions … a 100F, 20% relative
humidity condition and a 69F, 100% relative humidity condition. If an engine were
run at the 100F point, and water was introduced and completely evaporated in the
inlet, the temperature would drop to 69F with a corresponding increase in relative
humidity.26 Extracting the specific humidity numbers from the graph below, one will
find a 0.71% water to air ratio increase.
110 grams − 60 grams
7005 grams / lb .
• 100 = 0 . 71 %
69F 100% RH Case
100F
20% RH
Case
Figure 4.2. Evaporating water will reduce air temperature from 100F to 69F and
26
increase relative humidity from 20% to 100%
NASA/CR—2004-212957
26
The engine performance impact on these two operating conditions can be
seen in Figure 4.3. By injecting 0.71% water to air ratio and keeping the engine
throttle setting unchanged (i.e. take off power setting or PS=50), the thrust
increases 7.7% and SFC increases 0.76%. NOx decreases 14.9% and T3
decreases 43R. When increasing the aircraft speed to the point of lift off (i.e. 0.25
Mach), the thrust further increases 9.07% more than the non-water misted engine.
SFC decreases 0.56% while NOx and T3 remain essentially unchanged from the
static condition.
This shows that water misting the engine inlet improves thrust, T3 and NOx
emissions. Next, since the study is only considering using water misting for NOx
reduction and not power increases, the throttle setting of the engine will be reduced
while water misting to keep the same thrust level as without water misting. Figure
4.4 now compares the previous data point of the 0.25 Mach condition (constant
throttle setting) to that of a reduced throttle setting, keeping the same thrust output
as the non-water misted condition.
Figure 4.4 shows that when the engine throttle is retarded to keep the same
thrust output (at 0.25 Mach) as without water misting, further improvements in SFC,
T3, T4 and NOx emissions are gained as compared to keeping the throttle setting
constant. When using a 0.71% water misting rate on a 100F day with 20% RH, the
engine’s SFC improves 3.25%, T3 decreases 88R, T4 decreases 163R and NOx
decreases 28%.
15%
0
Aero Engine SLS
Aero Engine 0.25 Mach
10%
-10
5%
-20
Thrust
SFC
T3 (Delta R)
EINOx
0%
-30
-5%
-40
-10%
-50
dld03-17.xls
-15%
T3
69F, 100RH
vs.
100F, 20 RH
(Equiv. to 0.71% water injection
-20%
-60
Figure 4.3. Evaporating water in inlet increases thrust, reduces NOx
and T3 with little impact on SFC
NASA/CR—2004-212957
27
15%
0
Aero Engine @ 0.25 Mach
-40
Same with no thrust increase
0%
-60
Thrust
-5%
-80
SFC
T3
-10%
-100
-15%
-120
-20%
-140
69F, 100RH
vs.
100F, 20 RH
(Equiv. to 0.71% water injection
-25%
-160
T4
EINOx
-30%
Temperature (Delta R)
5%
-20
-180
DLD03-17.xls
10%
Figure 4.4. Retarding the throttle to keep constant thrust while using LPC water
injection further reduces SFC, NOx, T3 and T4
SFC
Increased
thrust with
increased SFC
Same
thrust with
reduced SFC
Non-water misted
100F, 20% RH
DLD03-46.xls
0.01
Water misted inlet
Equiv. to 69F, 100% RH
0
10,000
20,000
30,000
40,000
50,000
60,000
70,000
80,000
90,000
Thrust (lb)
Figure 4.5. Water misting during takeoff either reduces SFC, or increases thrust
From the previous two charts, it appears that by using water misting in the
engine inlet, a thrust increase may be gained or SFC can be improved by moving to
a new operating line. Figure 4.5 illustrates this relationship.
NASA/CR—2004-212957
28
5.0%
-10
4.5%
4.0%
-30
Aero Engine (100F day)
SFC (% change)
3.5%
3.0%
-50
Aero Engine (78F day)
(0.50 water to air ratio)
-70
2.5%
-90
2.0%
1.5%
-110
T4 Decrease (Delta R)
(0.50 water to air ratio)
-130
0.5%
-150
0.0%
dld03-17.xls
1.0%
T4
SFC
Figure 4.6. Starting temperature doesn’t make much difference on SFC or T4
reduction as long as injected water can completely evaporate
Figure 4.6 shows a lesser water injection rate of 50% water to air ratio from
two starting temperatures, 100F and 78F. It illustrates that there is not a strong
dependency on the starting temperature for SFC and T4 improvements. Thus, as
long as the water can be completely evaporated to reduce the air temperature in
the inlet and the air saturation point remains less than 100%, engine performance
improvements can be had.
For the following LPC, HPC and combustor injections methods, the NASA
engine performance program was used to estimate the affect of water injection on
the engine. Water is only injected into the engine core in these scenarios.
4.2.2 LP Compressor Injection
Current industrial engines use a Low Pressure Compressor (LPC) water
misting injection rate of approximately 0.5% to 0.87% water to core air flow ratio on
90F days. This resulted in a small NOx improvement. To increase the NOx
reduction level, the water flow rate should be increased. When the rate is
increased to 2.2%, the NOx reduction potential is estimated to be about 50%.44
This injection rate should be achievable and may be able to reach levels as high as
3%.45
Figure 4.7 compares the data points discussed in section 4.1 from the
industrial engine and section 4.2.1 from the aero engine to that of injecting water
directly into the LPC and increasing the water flow rate to 2.2%. It shows that when
keeping thrust constant, a 3.51% decrease in SFC will be obtained, a 46.5% NOx
reduction and large 436R temperature reduction in T4 will be experienced over a
non-water misted engine.
NASA/CR—2004-212957
29
0
20%
50
Reduce throttle to keep thrust constant
for case #4
100
0%
-10%
150
Thrust
or
Power
SFC
200
250
-20%
1) Industrial Engine (GE PS)
300
(0.53 water to air ratio)
-30%
2) Aero Engine SLS (Boeing)
350
(0.71% water to air ratio)
T4 Decrease (Delta R)
10%
3) Aero Engine 0.25 Mach (Boeing)
-40%
(2.2% water to air ratio & no thrust gain)
450
T4
EINOx
-50%
500
Figure 4.7. Increasing water/air ratio to 2.2% further reduces NOx, T4 and SFC
-10%
LPC
Injected
T4
0
-50
HPC
Injected
-100
-15%
-150
-20%
-200
-25%
-250
-30%
-300
-35%
-350
-40%
-400
-45%
-450
-50%
-500
Delta T4 (Deg R)
-5%
NOx
SFC
0%
Figure 4.8. HPC injection offers less SFC, NOx and T4 benefit than LPC injection
4.2.3 HP Compressor Injection
Injection of atomized water after the LPC and before the HPC, results in less
of a performance improvement than before the LPC. As shown in Figure 4.8, SFC
only improves 1.72% for HPC injection instead of 3.51% for the LPC case, NOx
decreases 44% instead of 47% and T4 decreases 335 instead of 436 deg R.
NASA/CR—2004-212957
30
DLD03-17.xls
400
(0.71% water to air ratio)
4) Aero Engine (NASA Glenn)
4.2.4 Combined LP and HP compressor injection
In the event of freezing conditions, it may be preferable to inject water directly
into the HPC instead of the LPC to avoid freezing of the water. However, as the
LPC injection method shows a better SFC performance benefit than the HPC
injection method, it would be preferable to normally inject water into the LPC.
4.2.5 Combustor Injection
Traditional combustor water injection systems have the advantage that, for a
given NOx reduction, they require much less water to be injected than a LPC or
HPC injected system. This is shown in Figure 4.9.
One of the disadvantages of a combustor injected system is the thermal
efficiency loss of the engine. In this system, the injected water partially quenches
the combustor flame temperature which leads to a reduction in pressure and
eventual thermodynamic efficiency. This system also looses the advantage of
improving compressor mass flow to offset the thermal loss as in the LPC system.
Figure 4.10 shows the relationship that as water injection rate into the combustor is
increased, NOx and thermal efficiency are both reduced, but power can be
increased by increasing the fuel flow rate.
Figure 4.9. Combustor water injection requires less water than LPC injection11
NASA/CR—2004-212957
31
Figure 4.10. Thermal efficiency and NOx decreases as water injection rate
increases46
Figure 4.11 shows these relationships as modeled in the NASA engine
performance program. For a water to fuel ratio of 0.5:1 on a standard day while
keeping power constant, the combustor water injected engine will experience an
adverse 2.02% increase in SFC. The engine will achieve a 50% NOx reduction, a
81R T4 decrease and unchanged T3. This is because only the turbine sees the
cooling effect of the water injection.
NASA/CR—2004-212957
32
10%
0
T3
-10
Thrust
SFC
-20
-10%
-30
-20%
-40
-30%
-50
-60
-40%
Delta Temperature (Deg R)
0%
-50%
-60%
-80
EINOx
SLS, 59F day,
0.5 to 1.0 water to fuel ratio
T4
-90
dld03-17.xls
-70
Figure 4.11. Injecting water into the combustor increases SFC while decreasing
NOx and T4
4.2.6 System Comparison
Table 4.1 shows the engine summary data for the Baseline engine, LPC, HPC
and combustor injection systems using the NASA NEPP. From a fuel efficiency
perspective, the LPC injection system is the preferred option.
Table 4.1, NASA NPSS engine performance summary
ALT Amb. Temp MACH Thrust (lb) T3 (deg R) T4
P3 (psia) SFC
EINOx
Baseline
0
529
0.25
74445
1636.5
3285 621.8
0.3794
58.78
LPC injection
0
529
0.25
74445
1410.2
2849 655.0
0.3661
31.43
HPC injection
0
529
0.25
74445
1427.9
2950 636.3
0.3729
32.69
Combustor inj. 0
529
0.25
74445
1635.6
3204 622.4
0.3870
-
4.3 Airframe System Description
Section 4.2 showed that the combustor injected system was nearly twice as
effective in reducing NOx and so required about half the water. Thus, two airframe
systems were designed … one for the combustor injection and one for the
compressor injection system.
4.3.1 Airframe system for combustor Injection
For a water to fuel ratio of 0.5:1 to achieve roughly a 50% reduction in NOx
and using standard times in mode for takeoff/climbout (section 2.2.3) and fuel
consumption rates for a large engine (section 4.4.5), a calculated water
consumption rate provides for a water tank capacity of 135 gallons for takeoff and
climbout conditions. Figure 4.12 shows the airframe system layout.
NASA/CR—2004-212957
33
Remote Service
Panel
Quantity
Indication
Manual Tank
Select Fill
Valve
Fill Connection
Drain Valve
Light
Water Tank
(135 Gallons Water)
Tank Quantity
Transducer
Tank
Fill Line
Drain
Valve
High Pressure
Pump (22,700 pph, 1K psig)
t
v
#2
Engine
Flow
Regulator
p
Shutoff
Valve
Line to
Engine
No. 2
Line to
Engine No. 1
Drain
Mast
Manifold
Drain
Valve
Anti
Siphon
Valve
Front
Spar
(Ref.)
Flow Pressure
Switch (2)
Figure 4.12. Airframe water system for direct combustion injection system
Remote Service
Panel
Quantity
Indication
Manual Tank
Select Fill
Valve
Fill Connection
Drain Valve
Light
Drain Valve (2)
Integral Water Tank
150 Gallons Water (2)
High Pressure Pump
(25,400 pph, 1K psig)
v
p
Tank
Fill Lines
t
Line to
Engine No. 1
v
p
Manifold
Drain
Valves (2)
Flow Pressure
Switch (2)
#2
Engine
Tank Quantity
Transducer (2)
Flow
Regulator
t
Line to
Engine
No. 2
Drain
Mast
Shutoff
Valve
Front
Spar
(Ref.)
Anti
Siphon
Valve
(2)
Figure 4.13. Airframe water system for LP compressor injection
The advantage of this system is that it uses one high pressure pump and has
only one drain mast. The disadvantage is that it requires a single, centrally
mounted water tank.
4.3.2 Airframe system for LP compressor injection
For the LP compressor injected system, more water is required to achieve the
same NOx reduction as the combustor system. Using the assumed 2.2% water to
core air flow ratio, and the standard time in mode of section 2.2.3, a water tank
capacity of 300 gallons was estimated. It uses two high pressure (534-750 psig)
pumps each capable of a 26,000 PPH flow rate. Figure 4.13 shows the layout of
such a system.
The system uses two water tanks, each located in the forward part of the wing
as shown in Figure 4.14. For safety reasons, there are areas in the wing near the
NASA/CR—2004-212957
34
engines that do not contain fuel. In the event of a catastrophic engine failure (e.g.
rotor burst), the areas around the engine where debris might penetrate the structure
or wing are kept free of fuel. These areas are called “dry bays” and are ideally
suited to house the water tanks as shown below.
The dry bay available area in each wing is capable of holding 407 gallons of
water. However, each water tank will be designed to hold 150 gallons of water that
will be sufficient to supply the engine with water to at least 3,000 feet altitude. In
both designs (4.3.1 and 4.3.2), there is a centrally located water fill and control
panel that is ground accessible.
Ri
b1
3
Ri
b8
The water tanks should be filled each time the airplane lands and not carry
water to the destination as the water would freeze in flight inside the tanks. In older
systems, the water lines and water dump mast were heated so that water could be
jettisoned in the event the water was not used during takeoff and also to drain the
lines of any remaining water.
Figure 4.14. 150 gallon tank located in each wing dry bay
NASA/CR—2004-212957
35
4.4 System Performance Summary
4.4.1 Weights
4.4.1.1 Airframe and Engine – The weight of the water misting system and
combustor injection systems for the airframe and engine are estimated to be no
greater than 360 lbs. This was determined from the weight of a 747-200 water
injection system. Weight improvements could be made over the 747 system.
Areas of improvement are:
1)
Water bladders in the composite wing could possibly be replaced with
a sealant type coating that would be applied to the inside of the
integral water tanks within the wing.
2)
The 36 KVA electric motors could possibly be down-sized. The
combustor injection system needs less water flow. For the LPC
system the motor and pump for the 747 provides the same required
water flow rate. However, if HP compressor bleed air could be used
to assist in the atomization of the water at the engine injection point,
lessening the required pressure (534-750 psig) from the water pumps,
then the power required for the pumps could be lessened which would
lead to lighter pumps.
3)
Improved technology. The 747 water injection system was designed
over 30 years ago. Weight improvements in components may have
been achieved during that time.
4.4.1.2 Water weight – The water weight for the combustor injection system is
1,127 lbs. The water weight of the LPC injection system is 2,505 lbs. (300 gallons
at 8.35 lb/gal). 30% of the LPC system water weight is consumed during the
takeoff roll and the rest consumed during the initial climb out period.
4.4.2 Thrust
Engine thrust is primarily a function of the mass and acceleration of the
gasses exiting the fan and engine core nozzles. This is the familiar thrust equation
Thrust (F) = mass (m) times acceleration (a). By adding water to the engine, the
mass flow is increased and so thrust increases.
For this study, it is assumed that the additional available thrust from water
injection will not be used. This avoids any safety related issues if there was a
failure with the system. Although water injection would enable a smaller, lighter
weight engine to be used (this would improve fuel efficiency), the initial cruise
altitude capability of the aircraft could be sacrificed. The engine size is determined
by the takeoff and cruise altitude capability.
4.4.3 Takeoff, Climb and Range Performance
During takeoff and climbout, because SFC is affected by the type of injection
system used, the combustor injected system will use 51 lbs. more fuel while the
LPC system will use 90 lbs. less fuel.
NASA/CR—2004-212957
36
5,000
637,000
4,500
636,000
30% of water used during takeoff roll
4,000
Altitude
634,000
2,500
633,000
2,000
Weight
1,500
632,000
Water runs out at this point
1,000
631,000
500
0
630,000
0
2
4
6
8
10
12
14
16
Distance (st miles)
Figure 4.15. 30% of water (750 lb. for LPC system) is used during takeoff roll, so
climb performance is minimally affected by weight
Both systems will use approximately 30% of the water during the takeoff roll.
All of the water will be consumed by the time the aircraft reaches 3,560 feet. Figure
4.15 shows the how the weight of the aircraft decreases throughout the climb
sequence by using both water and fuel.
For aircraft that are range limited by the amount of fuel that can be carried,
there is a slight range penalty of 7 nmi. for carrying the water injection systems. For
aircraft that are limited by the take off gross weight of the aircraft, carrying the water
injection system and the 2,505 lb. of water will replace that same amount (weight)
of jet fuel. Thus, the aircraft range is reduced by 80 nmi.
A positive variable for the system is the reduction in T4 when using water
injection. Figure 4.16 shows that the highest turbine inlet temperature occurs
during the takeoff roll and up to the point right after takeoff where the throttles are
reduced (cutback). When using LPC water injection, Section 4.2.2 showed a 436 R
decrease in T4. This reduction in T4 is shown in Figure 4.16 by the dashed line.
Thus, peak temperatures that the engine will experience with water injection would
now be at the top of climb and the very highest it will ever see are reduced by more
than 200 R from the previous takeoff peak value. This would no doubt improve the
life of the current engine turbine inlet nozzle guide vanes and HP turbine blades.
NASA/CR—2004-212957
37
636TOGW_LSPS.xls
Altitude (ft)
3,000
Airplane Weight (lbs.)
635,000
3,500
3,400
3,300
Turbine Inlet Temp (R)
3,200
Without water injection
3,100
3,000
2,900
2,800
2,700
With LPC water misting
H20_climb.xls
2,600
2,500
2,400
0
5,000
10,000
15,000
20,000
25,000
30,000
35,000
Altitude (ft)
Figure 4.16. Water misting greatly reduces T4 during the most critical phase of
flight on current technology engines.
However, as future aircraft engines increase their fan by pass ratios, this may
lead to decreases in T4 levels at takeoff and increased levels at top of climb. Thus,
the temperature reduction benefit might not be as great for future engines than for
current technology engines.
4.4.4 Fuel Use
Very little fuel use impact will be experienced by the aircraft, either during the
take takeoff, climbout, or the cruise parts of the flight. However, the following
documents these small changes.
For the combustor water injected engine, a 2% thermal efficiency loss is
experienced during takeoff. This results in a 51 pound (7.6 gallon) fuel use penalty.
For the LPC injected system, a 3.51% SFC improvement is anticipated under
standard day (non-freezing) conditions. This results in a 90 pound (13.4 gallon) fuel
savings. Figure 4.17 shows the changes in fuel use for the baseline engine as well
as the engines with combustor and LPC injection methods.
NASA/CR—2004-212957
38
51 lbs. more fuel
for combustor
injection
3,000
2,500
2,000
LPC Injection on
std. day
1,500
1,000
Non-water injected
or
LPC Injection on Cold Day
500
90 lbs.
less fuel
for LPC
injection
636TOGW_LSPS.xls
Fuel Used (lbs)
Combustor Injection
Water injection to
3,560 ft. altitude
0
0
2
4
6
8
10
12
14
Distance from Takeoff (st miles)
Figure 4.17. Combustor water injection uses 51 lb. more fuel from takeoff to 3,560’
altitude while LPC injection uses 90 lb. less fuel than base engine on standard day
Water
Figure 4.18. Water wash is used to clean engine and restore performance
During cruise, any weight increase of the airplane will require additional fuel.
However, as this system is only expected to weigh less than 360 lbs, a very small
fuel use penalty will be experienced. For the study airplane on a 3,000 nmi
mission, a 63 lb (9.3 gallon) fuel use penalty can be expected.
When using water misting injection into the LPC for takeoff, some cleaning of
the compressor may occur.47 Presently some aircraft operators use engine water
washing during maintenance periods to clean the compressor and turbine sections
of the engine to restore engine performance (Figure 4.18).
NASA/CR—2004-212957
39
The amount of performance improvement from water wash is not well
documented.
On 747 aircraft, average performance deterioration, from 3 engine
manufacturers, reaches the 3-4% level after several years (Figure 4.19). This
deterioration is not only from dirty compressor and turbine blades, but mechanical
deterioration as well. When water washing an engine, it is generally believed to
contribute a 0.5 – 1.0% SFC restoration. If this level of restoration were achieved, it
would result in a 294,263 to 588,525 lb. fuel savings per airplane per year.
However, as this benefit is speculative with using water misting injection, it will not
be included in the study.
4.4.5 Emissions
NOx Emissions levels are typically referenced to Emissions Indices or EINOx.
This is the emissions level (grams) divided by the fuel use (kg). Standard ICAO
emissions databases list the EINOx and fuel use numbers for takeoff, climbout,
approach and idle conditions. Figure 4.20 shows an example of such a data
sheet.48 For this exercise, ICAO NOx emissions for a GE90-85B engine (ID
2GE064) was obtained and used to validate the NASA NEPP emissions data. As
the NEPP data is intended to simulate a generic GE90 or PW4000 engine, the
emissions reduction potential is similar to either engine. Appendix A shows more
on the calculation methodology and matching of results to Boeing predicted and
actual engine data points.
All 747-400 Aircraft
Average of all engine manufacturers
Change in Fuel Mileage (% from new)
2
1
0
-1
+1 Sigma
-2
Mean
Deterioration
-3
-1 Sigma
Estimated
Trend
-4
-5
0
1
2
3
4
5
6
Years in Service
Figure 4.19. Deterioration can reduce SFC
NASA/CR—2004-212957
40
7
DD98-43
Figure 4.20. ICAO emissions data from the GE90-85 engine (2GE064) was used to
validate the NASA NEPP emissions results50
4.4.5.1 Taxi Emissions – The amount of time the aircraft is taxiing, and the
emissions index, both have a large affect on the total amount of emissions
generated. Although the engine is operating for a long period of time during taxi
(26 minutes), it has a very low NOx emissions index (grams of NOx per kg of fuel
burned) and therefore contributes a small amount of NOx during the LTO cycle.
Using the GE90-85B emissions database above, the idle portion contributes only
0.64% of the cycle NOx. Conversely, CO and HC emissions have high levels at this
low power setting, contributing 34% and 35% of the LTO CO and HC emissions.
Other engines typically have even high contribution percentages of CO and HC at
idle conditions. Figure 4.21 shows the relative emissions contributions for the
various phases of the LTO cycle. Thus, using water injection during the taxi phase
of the LTO would have little impact on NOx, but increase CO emissions. For this
reason water injection was only considered for takeoff and climb-out conditions.
NASA/CR—2004-212957
41
Approach
90%
Climb
80%
70%
60%
50%
40%
Takeoff
30%
20%
10%
LTO_emiss.xls
Landing Take Off Emissions (%)
100%
Idle
0%
NOx (%)
CO (%)
HC (%)
Figure 4.21. As little NOx emissions are generated during taxi, water injection was
not used for this phase
4.4.5.2 Takeoff Emissions – The takeoff and climbout portions of the LTO
cycle contribute the most to an airplane’s NOx emissions. As NOx is the emissions
of focus at airports, water injection would be able to help achieve reductions in local
airport emissions by using it from the moment takeoff power is commanded to the
point of water exhaustion which occurs outside of the airport boundary.
The governing mechanism in water injection is the lowering of the
49
stoichiometric flame temperature. This tends to be strictly a thermal phenomenon
and typically lowers prompt NOx a very small amount, which is a small contributor
to overall NOx production.
The primary zone stoichiometry has an effect on the effectiveness of water in
reducing emissions.50 Fuel-rich primary zones being more susceptible to
improvements in NOx reduction with water injection.
Using the NASA NEPP results and Boeing airplane performance decks for
validation, Figure 4.22 shows the altitude versus distance profile for the study
airplane. In addition, it also shows the standard NOx generation profile and the
reduced LPC water misted NOx generation. At 3,560 feet, the 300 gallons of water
will have been exhausted. At this point, the amount of NOx reduction will have
been 49.2 lb of NOx, achieving a 46.5% reduction in takeoff and climbout NOx.
NASA/CR—2004-212957
42
5,000
60000
4,500
50000
4,000
40000
49.2 lb. (22.3 kg)
NOx saved
Normal NOx
2,500
30000
2,000
Water runs out at
this point
1,500
20000
Water Injected NOx
1,000
10000
500
0
0
0
2
4
6
8
10
12
14
16
18
Distance (st miles)
636TOGW_LSPS.xls
Altitude (ft)
3,000
Airplane NOx (grams)
Altitude
3,500
Figure 4.22. NOx is reduced 46.5% during takeoff and climb-out saving 49 lbs. of
NOx emissions to 3,560 feet (beyond normal LTO cycle altitude)
The NOx emissions reduction during this specific takeoff procedure, and
airplane configuration, is more than what would be achieved when simply
calculating a 46.5% NOx reduction when using the ICAO LTO cycle for takeoff.
The 49 lbs. NOx reduction occurred when using 300 gallons of water, which took
the airplane to an altitude of 3,560 ft and water misting time interval of 3.3 minutes
versus the typical 3,000 ft and 2.9 minutes for the LTO cycle. In addition, fuel
savings are realized in our calculations which further reduces total NOx emissions.
Everyday NOx savings for aircraft in use would most likely be lower since they
typically operate in the 70% load factor range instead of the study’s 100% load
factor and 100% engine thrust.
The amount of NOx savings is also dependant on the type of engine used.
Smaller, lower pressure ratio engines would have less reduction potential while
larger, higher pressure ratio engines would achieve more. For example, the GE9085B engine, which closely resembles the study engine, has a total LTO NOx
emissions rate of 108 lbs. per LTO cycle. The smallest 777 engine is the PW4077
that has a NOx emissions rate of 63 lbs per LTO while a larger PW4098 engine has
a level of 142 lbs.
The study engine, and 49 lb. NOx reduction calculation, represents an
average sized engine for the 777 with a reasonable reduction potential level.
Releasing exact reduction levels would involve disclosure of engine company
proprietary data.
NASA/CR—2004-212957
43
4.4.5.2 HC and CO Emissions Tradeoffs
Typically, there are tradeoffs required in order to achieve this level of NOx
reduction. One of the design philosophies behind low NOx combustors is to wellmix the fuel-air mixture prior to combustion and achieve a more homogenous
process that reduces hot burning zones within the combustor. However, as these
high temperature zones are reduced by leaning the fuel-air mixture, CO and HC
emissions tend to rise as illustrated in Figure 4.23.
Indeed this relationship also exists for the water misting intercooler system.
As water is injected to reduce NOx, CO and UHC climb as shown in Figure 4.24.
Emissions
NOx
HC
D LD 01- 34
CO
Smoke
- Fuel/Air Mixture Lean
Ideal
Rich
- Flame Temperature Cool
Hot
Cool
Figure 4.23. Emissions Relationship
NASA/CR—2004-212957
44
Figure 4.24. HC and CO increase with decreasing NOx54
Table 4.2. Aeroderivative engines used in Figure 4.25
Engine Model
LM2500
LM6000
Aero Engine Parent
CF6-6
CF6-80C2
Power (shaft HP)
31,200 to 42,000
56,795 to 58,932
Aero certification date
1970
1985
Combustor
Annular with 30 fuel nozzles
Annular with 30 fuel nozzles
Pressure Ratio
18:1
29:1
NOx (ppmvd, ref. 15% O2) on
316
403
42
42
distillate fuel
NOx (with water injection)
The rate at which CO and HC emissions climb is dependant on the engine
and, no doubt, the engine operating conditions. Figure 4.25 show a difference in
the CO production rate between two GE industrial engines, a LM2500 and LM6000,
that are further described in Table 4.2.
NASA/CR—2004-212957
45
The LM6000 engine shows practically no increase in CO until a water to fuel
ratio of 1.4 is reached. However, the LM2500 engine shows CO increases as soon
51
as water is introduced into the combustor. This difference in behavior may be due
to the engine combustor operating characteristics. Namely, the LM2500 engine’s
relatively low pressure ratio of 18:1 versus the newer LM6000 engine’s pressure
ratio of 29:1. Thus, the LM6000 engine is operating at a significantly higher
combustor pressure and therefore temperature, both of which tend to reduce CO
emissions.
As the LPC study engine is operating at an approximate equivalent 0.5 water
to fuel ratio, and the combustor operating conditions are more in line with the
LM6000 engine, the CO level may remain essentially unchanged. However, this
has yet to be proven and it could even increase some 50% if it follows the lower
pressure ratio LM2500 engine trend.
Generally, smoke has been reported to reduce when using water injection.
Figure 4.26 shows that, for a combustor water injected engine using Jet-A fuel,
smoke is reduced as the water rate is increased. No reported results on smoke
emissions were found for LPC injected engines.
Low OPR (18:1)
older technology
CF6 derivative
engine
High OPR (29:1)
current
technology CF6
derivative engine
Operation Range?
Figure 4.25. CO generation by water injection is dependant on engine type16
NASA/CR—2004-212957
46
Figure 4.26. Test results show that smoke may decrease with water injection
4
4.4.6 Noise
When water is added to the core of the engine, the total engine core flow
density increases. Since a constant thrust is being maintained in the takeoff cycle,
the velocity of the core needs to be reduced to compensate for the increased mass
flow. For this particular engine cycle, the fan flow mass and velocity decreased to
maintain the same thrust level. The mass flows and velocity of the core and fan
flow will determine the noise level of the engine. Figure 4.27 shows that as the
engine core mass flow increases with the addition of water, the core velocity is
reduced as well as the fan mass flow and velocity. Together, these averaged flows
result in a 0.61 db reduction in engine noise. This noise benefit may be lost if the
engines were resized to take advantage of the added available thrust.
NASA/CR—2004-212957
47
Core
15%
Percent Change
10%
5%
Fan
Noise
0%
mass
vel
mass
vel
-0.61 dba
-10%
water_inj_noise.xls
-5%
Figure 4.27. Noise decreases slightly because mass averaged jet velocity
decreases
4.4.7 Maintenance
Due to the large decreases in turbine inlet temperature documented in
Sections 4.2 and 4.4 of this report, it is anticipated that increased hot section life will
be achieved in current technology engines when using water misting intercooler
technology. This could have a large impact on reduced costs for newer aircraft
engines. However, as this cost savings is not easily calculated and data are
currently being collected, it is not included in this study.
Historical water injection system and engine maintenance costs for the 747100 and 747–200 aircraft were reported in section 2.4.1 to be $20.62 (2003 dollars)
per takeoff. As the 777 has two engines as compared to 4 engines for the 747, this
cost will be reduced to $10.31 per takeoff.
Other maintenance concerns and comments from Boeing and engine
company reviewers are listed in Appendix C.
4.4.8 Water Conditioning and Cost
According to airport industry agreed upon service costs,
demineralized/conditioned water (includes transportation to the aircraft) costs are
approximately $23.59 per airplane service. About half of this cost ($12.28) is for
the tankering of the water to the airplane. This leaves $11.31 for the conditioned
water cost, or $0.038 per gallon. For production of water at the airport, this water
cost would decrease to at least $0.026 per gallon (Appendix B) or could even go
down to as low as roughly $0.01/gallon (Appendix C) for an optimized system.
NASA/CR—2004-212957
48
The airplane water service cost is derived in the following manner … airports
agree to provide services according to mutual assistance ground service
agreements. Table 4.3 shows the average rate per hour for providing conditioned
water to the aircraft. A typical 600 gallon ground service vehicle52 was assumed to
be used for the study. As the study airplane uses 300 gallons of water, the water
truck would be able to service two large aircraft per hour ($47.12 ÷ 2 services =
$23.59 per service).
Table 4.3, Water conditioning and delivery costs53
Seattle
$43.54
Anchorage
$50.07
Honolulu
$47.89
Average
$47.12 per hour or $23.59 per airplane
Appendix B lists an alternate water cost calculation methodology where water
is mass-produced at the airport using commercial water conditioning equipment. In
this scenario, the water costs for an industrial gas turbine engine were used.54 For
a large airport, such as Seattle-Tacoma, providing demineralized water to all
commercial aircraft (non-regional jet), the water purification system would need to
provide about 11,290,835 gallons per year. The water cost for such a system
would be $0.026 per gallon. Transportation cost is the major expense, costing
$12.28 per aircraft. For a 300 passenger aircraft this would result in a water cost of
$20.08 (300 gallons x $0.026 + $12.28).
4.4.9 System Cost
The additional system cost is estimated to be between $100,000 and
$200,000 for each aircraft. This does not include non-recurring engineering costs.
On a simple straight-line basis, the non-recurring costs are estimated to add
approximately $9,200 per aircraft.
NASA/CR—2004-212957
49
4.4.10 Operating Economics
Table 4.4 shows the anticipated performance impacts on the aircraft for using
combustor injection as well as LPC injection, both on a water saturated or cold
(32F) day and standard day (59F) conditions.
Table 4.4. Performance Impacts for 777-type Airplane
Combustor
Injection
LPC Inj.
(cold day)
LPC Inj.
(standard day)
Mission length (nmi)
3,000
3,000
3,000
Trips per year
475
475
475
Incremental T/O fuel
(lb./trip)
51
-56
-90
Incremental cruise fuel
(lb./ trip)
63
63
63
Water used
(gallon)
132
300
300
Range loss at 100% LF
(nmi for MTOW limited) 80
(nmi for fuel tank limited) 7
80
7
80
7
Capital Costs
$159,202
$159,202
$159,202
NOx reduction per LTO
(lbs)
52.9
47.2
49.2
NASA/CR—2004-212957
50
Table 4.5. Fuel, Water and Maintenance Costs for 777-type Airplane
Combustor
Injection
LPC Inj.
(cold day)
LPC Inj.
(ave. day)
∆Fuel cost per departure
$5.45
$-5.98
-$9.61
∆Fuel cost for cruise
(@ $0.72/gallon)
6.73
6.73
6.73
Water cost @ 0.026$/gal
(per departure)
3.43
7.80
7.80
Water service cost
($ per departure)
12.28
12.28
12.28
∆Maintenance per departure 10.31
10.31
10.31
Simple capital cost
13.41
(25 year life, 475 trips/yr)
13.41
13.41
Total ∆ cost per departure
$44.55
$40.92
$51.61
Table 4.6. Water misting NOx reduction cost
Combustor
Injection
LPC Inj.
(cold day)
LPC inj.
(ave. day)
Total ∆ cost per departure
51.61
44.55
40.92
NOx emissions reduction
(lbs. per LTO)
52.9
47.16
49.2
$1,951
2.56
1,889
2.48
1,663
2.18
Emissions reduction cost
($/ton)
($EU/kg)
NASA/CR—2004-212957
51
10,000
Cost per
Departure
8,000
6,000
4,000
Capital
Cost
$13.41
Maintenance
$10.31
Truck
Service
$12.28
Water Cost
2,000
3K nmi Cruise
Fuel Penalty
360 lbs.
0
777 with Water
Injection*
$7.80
$6.73
dld99-10.xls
NOx Cost ($/ton/mission)
12,000
* includes $9.61 takeoff fuel savings
Figure 4.27. Cost breakdown of airplane water misting intercooler system.
Figure 4.27 shows the breakdown in water misting costs per takeoff and also
shows the cost/benefit of the technology.
Figure 4.28 illustrates how the cost of the airplane engine water misting
intercooler system compares with the costs that industry typically is paying for NOx
reduction through various emissions reduction technologies. The chart also lists
the cost/benefit ratio that airplane operators are subject to at Swedish airports by
emissions-based landing fees. Thus, for the study airplane, the emissions
cost/benefit ratio is very favorable. Other sized airplanes/engines will have different
ratios, and in some cases (e.g. small airplanes with frequent stops) the cost/benefit
ratio may be substantially worse.
NASA/CR—2004-212957
52
60,000
Emissions Costs ($/ton)
50,000
40,000
30,000
20,000
10,000
1,663
0
EU Urban
Swedish 777
emissions fee
So. California
EU Rural
US National
NOx trading 777 water
credit(1)
misting (2)
Notes: 1) Typical emissions Trading Credit (RTC) in 1999.
2) 777-200ER on 3,000 nmi mission
Figure 4.28. Large airplane engine water misting may prove to be substantially less
costly than other industrial NOx reduction technologies.
4. 4.11 Airline Operator Survey
Nine airlines were surveyed about the use of water injection for NOx
reduction. Only 2 responded. Both airlines indicated they had used water injection
for their older 747 aircraft and both noted they had experienced added cost due to
the requirement of obtaining demineralized water. One operator noted the water
pumps would occasionally freeze. The other reported increased maintenance due
to corrosion inside the water system. Based on their previous experience, neither
airline welcomed the technology. However, one airline indicated that for a newly
designed system, the previous technical difficulties would most likely be avoided. If
hot section life could be increased, then they might be interested in the technology.
rd
A 3 airline that did not respond directly to the survey but indicated that the
technology would be of interest should airport emissions landing fees increase or
operating restrictions come into place.
NASA/CR—2004-212957
53
5.0 CONCLUSIONS AND RECOMMENDATIONS
5.1 Summary of Results
Water injection has been used on industrial gas turbine engines for about 30
years to reduce NOx emissions. Water injection was used on Boeing 707 aircraft
(circa 1958) and early 747 aircraft (circa 1969) to increase thrust. This current
study suggests that water injection can be used on new aircraft to reduce takeoff
NOx emissions at very competitive costs.
The study results showed that the newer water misting system, where
atomized water is injected before the LP compressor, has a beneficial effect on
engine performance and NOx emissions. This is particularly true for operations on
hot days. When water is injected prior to the HP compressor, NOx emissions are
equally reduced, but some fuel efficiency gains are lost. When water is injected
directly into the combustion chamber, one will find a much improved reduction in
NOx emissions for the same amount of water. However, engine fuel efficiency
suffers and turbine inlet temperature reduction is not as great as for the LPC
injected system.
Based on a 305 passenger future technology airplane with current technology
engines delivering 85,000 lb. thrust, an airplane operator would experience a 46.5%
NOx reduction and 3.5% SFC improvement on a 69F day during the takeoff and
climb-out portion of the mission. This will avoid contributing 49.2 lb. of NOx to the
airport environment and save 90 lb. of fuel during the 3.3 minute takeoff and climbout phase. 63 lb. of fuel would be required to carry the water misting intercooler
support equipment on a 3,000 nmi. mission. For takeoff during days when the air is
fully water saturated or below freezing conditions, NOx reduction is still achieved by
injecting water into the HPC, but the fuel efficiency improvements will be reduced to
1.7% as there will be no evaporative cooling experienced in the engine inlet and
LPC. For a direct combustor injection system, an operator would experience a
2.0% decrease in fuel efficiency during takeoff. This is compounded by a 63 lb. fuel
use penalty on a 3,000 nmi mission by having to carry 360 lb. of added equipment
for the water injection system. However, much greater (e.g. 70-85%) NOx
emissions reduction potential could be achieved by combustor injection.
The cost effectiveness of the LPC water misting system appears to outweigh
that of the other systems. From initial estimates, this system would cost the
operator an average of $40.92 per takeoff cycle for the 305 passenger study
airplane. About 32% of this cost is due to the capital cost of the equipment, 25%
for increased maintenance with the rest being accounted for by water and servicing
costs.
Other potential savings for airplane operators were suggested in the study,
though not analyzed, for improvement in engine hot section life. When using water
misting injection, the turbine inlet temperature is expected to decrease
NASA/CR—2004-212957
54
approximately 436 ºR during the most demanding part of the airplane mission -take off roll and initial climb-out. Industrial engines have experienced compressor
blade erosion when using water misting continuously. If this system were used just
for takeoff on aero engines, this erosion challenge may be turned into a cleaning
opportunity. Clean compressors and turbines have been shown to restore from ½
to 1% of SFC. Analysis and test work would be required to understand these
effects and validate the potential benefits.
5.2 Analysis of Results
European airports in England, Sweden and Switzerland are, or will be,
charging airport landing fees based on airplane LTO emissions. For a 777-200ER
airplane (636.5K MTOW) using a P&W non-water misted engine, the emissions
penalty portion of the landing fee in Sweden would be $537.00. A certain amount
of NOx emissions from each “clean” airplane is permitted at the airport. If one looks
at the penalty cost (i.e. $537.00) of the NOx emissions beyond the allowed “clean”
amount, this adds up to costing $52,455/ton for the additional NOx emitted. This is
substantially more than the estimated $1,663/ton cost for the water misting NOx
reduction technology studied in this report.
If the water misting injection system did indeed contribute to engine cleaning,
the ½-1% fuel savings would more than offset the water injection cost on missions
of 3,000 nmi or more.
Engine hot section components are very expensive and have shortened life
with very high operating temperatures. With the reduction in turbine operating
temperature during takeoff, there may be an opportunity to improve engine hot
section life and reduce operator costs. However, these potential cost savings were
not able to be quantified in the study.
As both of the water injection and water misting systems have the capability to
increase thrust, taking advantage of this potential during critical event episodes
(e.g. hot-day single engine-out) may have a safety benefit.
Although many questions remain, as evidenced by the feedback shown in
Appendix C, the study suggested the technology may be able to offer performance
and cost improvements over older style water injection. This could provide a very
cost competitive reduction in airport NOx emissions to enable the continued growth
of aviation.
5.3 Cost Uncertainties
An early draft of this high-level report generated much discussion amongst the
engine and airframe manufacturers. Many uncertainties, that could impact the
$1,663/ton NOx reduction cost/benefit ratio, were suggested. Among them are:
- The study did not reiterate the study airplane design to use water injection
only to 3,000 ft. (the 300 gallons of water ran out at 3,560 ft. in this study and gave
more NOx reduction than for a 3,000 ft. altitude design)
NASA/CR—2004-212957
55
- Engines with lower pressure ratios would benefit less from water misting.
- Smaller aircraft will probably have higher water misting operating costs.
- Combustor water injection will offer a greater degree of NOx reduction. A
tradeoff study with different water injection rates will highlight the best cost/benefit
ratio for each technology and determine which is the better option.
- As range is slightly reduced with the technology, a new airplane with higher
MTOW (to recover range) would suffer from increased landing fees.
- Increasing the number of takeoffs per year for large airplanes could reduce
the operating cost.
Given these uncertainties, the GE Aero Engine group has suggested that
costs could range from a low of $1,297/ton to a high of $8,380/ton of NOx reduced.
If the water service costs could be reduced, as suggested by the CH2M group,
these costs could be cut in half. Should the water wash benefits to SFC prove real,
the costs would be cut to nil. Further, T4 reduction benefits and resulting hot
section life improvements could conceivably turn water misting costs into savings
for the airlines.
5.4 Recommendation
As this study only offered a “quick look” at the cost and benefits of different
water injection methods, a more in-depth analysis is needed. Aerospace engine
companies, airplane manufacturers, NASA and government organizations together
need to further evaluate this technology and identify the steps necessary to bring
such technology to maturity for commercial airplane application.
Issues (both bad and good) that need to be addressed are listed: impact on
engine hot section life, engine high pressure compressor operability issues, impact
of the technology to smaller and larger aircraft/engines, range of cost/benefits due
to uncertainty, emissions prediction validations (including soot) with actual test
results, safety analysis, find the optimum balance between water misting rate and
NOx reduction (e.g. 50% or 85% NOx reduction), airplane range tradeoff issues,
possible compressor cleaning effects, water droplet size, LPC water misting vs.
combustor water injection architecture determination, evaluate system weight
reduction opportunities and lastly, water misting benefits to very high pressure ratio
engines (e.g. UEET powerplants with OPR of 60-70).
This NOx reduction technology appears to offer attractive enough benefits that
it merits further, more in depth investigation.
NASA/CR—2004-212957
56
APPENDIX A. NOX CALCULATION METHODOLOGY
The NASA NEPP code used an equation to predict the NOx emissions at any given
power setting, based on the engine’s T3 and P3 conditions (Equation 1 in section
2.2.1). The NASA NEPP data is not intended to replicate any particular engine, but
it does reflect anticipated GE90 and PW4000 types of engine performance. For the
combustor emissions data, the NASA NEPP NOx code attempted to mimic GE90type of engine performance. In order to validate the NASA equation, two steps
were taken; 1) the NASA predicted NOx was compared to ICAO engine test data
points, and 2) the NASA NOx equation was used with Boeing GE90-85B proprietary
engine performance data to predict NOx
For step one, the NASA emissions code closely predicted EINOx numbers for
the four ICAO LTO measurement points (takeoff, climbout, approach and idle.)
However, it does not confirm the under laying basis for the EINOx prediction -- T3
and P3 data points.
The next step was to see how closely the code predicts EINOx given that
actual engine performance points (i.e. T3, P3) are used. Using Boeing GE90-85B
engine performance data and the NASA NEPP EINOx equation, Figure A-1
illustrates the differences between the NASA EINOx equation prediction and actual
test data recorded in the ICAO data sheets. Results from the NASA NOx equation
are shown in the down-ward sloping line, given Boeing T3, P3 and thrust inputs for
the GE90-85 engine. The two ICAO data points for this engine (ID # 3GE064) are
shown for 100% (takeoff) and 85% (climb out) thrust levels. There is a 10.58
EINOx difference between takeoff and climb using ICAO data points. However, the
NASA equation predicts a 7.95 EINOx difference, underpredicting the NOx
reduction by 7.4% when at the 85% power setting and 35.7 EINOx point.
Therefore, when engine T3 levels are reduced while using water misting injection, a
greater NOx reduction may occur than was predicted by the NASA NEPP code.
For the purposes of this high-level study, these levels of NOx accuracy are
within reason. Further improvements in the emissions accuracy prediction would
require disclosure of proprietary engine data and is outside the scope of this study.
NASA/CR—2004-212957
57
50
45
10.58 - 7.95
40
35.7 EINOx
EINOx
35
X 100 = 7.4% under predict
NOx reduction for 10.58
GE90 engine
7.95
30
25
NASA NOx Equation
GE90-85_SFC_Std_SLS.xls
20
GE90-85B ICAO
Data Points
15
10
5
0
50%
Equivalent T3
point with water
injection
60%
65%
70%
75%
80%
85%
90%
95%
Thrust (% takeoff)
Figure A-1. Using Boeing GE90-85B engine data, the NASA NOx equation under
predicts NOx reduction potential by 7.4% at the 85% thrust level
NASA/CR—2004-212957
58
100%
APPENDIX B. COST ESTIMATION OF WATER INJECTION
Study Ground Rules
In this investigation, the overall water conditioning costs for the entire
commercial aviation fleet are estimated. Air traffic data for Seattle-Tacoma (SeaTac) international airport are used as a basis for the cost estimation.
The focus is on airplanes larger than 100 seats, which exclude regional
services and commuters (representing almost half of Sea-Tac’s air traffic volume.)
It is anticipated that water injection will yield little value for regional airplanes.
Seattle-Tacoma International Airport Traffic 2002
The investigation is based on annual 2002 statistics and also for the month
December 2002. Table B-1 shows all the Boeing, Airbus, McDonnell Douglas and
Lockheed L-1011 aircraft that are included in the airport’s statistics.
Figure B-1 illustrates a further breakdown of commercial aircraft, by type, for
non-regional airplanes. By far, the highest percentage of aircraft at Sea-Tac are
Boeing 737 aircraft.
Table B-1: Airplane Cycle Statistics SEA Int’l55
Dec/02
2002
Departures
9,247
113,910
Boeing
6,113
77,183
Airbus
1,187
12,203
6
101
1,941
24,423
Lockheed
MDD
Annual SEA-TAC Traffic 2002
*1000
60
50
40
30
20
10
A3
0
A3 0
1
A3 0
1
A3 9
2
A3 0
2
A3 1
40
B7
1
B7 7
2
B7 7
37
B7
3
B7 8
4
B7 7
4F
B7
5
B7 7
6
B7 7
7
D 7
C
1
DC0
D 8
C
L1 9
M 01
D
M 10
D1
M 1
M D8
D
M 82
D
M 83
D8
M 7
D
9
0
A/C Type
Figure B-1: Annual Traffic Seattle Tacoma International Airport 2002
NASA/CR—2004-212957
59
Figure B-2 shows the monthly traffic at Sea-Tac for December 2002. The
distribution is similar to the annual traffic shown in Figure B-1. Table B-2 shows the
tabular data that was used to generate Figures B-1 and B-2.
SEA-TAC Traffic Dec-02
*1000
5
4
3
2
1
A3
A300
A310
A319
A320
A321
B740
B717
B727
B737
B738
B7 47
4
B7 F
B757
B767
D 77
C1
DC0
DC8
L1 9
M 01
D
M 10
D1
M 1
M D8
D
M 82
D
M 83
D8
M 7
D
9
0
Figure B-2: Sea-Tac Traffic December 2002
Table B-2: Sea-Tac Traffic Statistics56
NASA/CR—2004-212957
60
In order to estimate the water requirements for all aircraft, the Boeing 777-200
is used as the basis for further calculations. Depending on thrust of both engines
for each airplane type, the required amount of water per aircraft type will be been
determined. Assuming the 777 has an average thrust rating of 390,000N (87,660
lb.) per power plant or 780,000N (175,320 lb.) per airplane and a water reservoir of
300 gallons, the water amount per Newton (lb. thrust) can be determined.
This is:
0.0003846 Gal/N, 0.3846 Gal/kN or
0.00171Gal/lb thrust
Based on these average values, the total amount of water required for each
airplane can be determined from its thrust. Figure B-3 shows the estimated amount
of water that would be required for each airplane type using the above
equation.
[Gal]
WTR/Cycle [GAL]
450
400
350
300
250
200
150
100
50
0
Engine Thrust acc. [N]
Water / Cycle [GAL]
Thrust [kN]
1200
1000
800
600
400
200
A3
A300
A310
A319
A320
A321
B740
B717
B727
B737
B738
B7 47
B74F
B757
B767
DC77
1
DC0
D 8
L1C9
M 01
MD10
D
M 11
M D8
D
M 82
D
M 83
D
Av M 87
er D9
ag
e
0
Figure B-3: Water required per cycle and thrust for various aircraft types
NASA/CR—2004-212957
61
Having determined the amount of water per cycle and knowing the total cycles
performed at Sea-Tac airport in 2002, the total required water for the airport is
calculated for the December 2002 traffic as well as for annual traffic in 2002. Using
the traffic data from Table B-2 and the water required data from Figure B-3, total
amount of water required for Sea-Tac airport is:
902,436 Gal / month [DEC 02] or
11,290,835 Gal / year [2002]
Table B-3 shows the breakdown by airplane type and also shows the monthly
and yearly average water consumption. This leads to an average daily requirement
of 30,022.3 Gal and a maximum of 33,036.8 Gal per day.
Table B-3: Water amount determination
NASA/CR—2004-212957
62
Water Quality
The quality of water used for water injection has to meet certain requirements
to ensure minimum maintenance and maximum durability of the engines. Especially
important is any water contamination with solids that can damage the engine
turbine material and severely reduce engine life cycles.
The following water quality criteria were specified from GE:57
•
Total Na, K, Pb, Va and Li ≤ 0.1 ppm,
•
Total dissolved and nondissolved solids < 5 ppm,
•
Total suspended solids ≤ 10 mg/Gal,
•
Maximum size of solids ≤ 20 µm (absolute).
Regarding water treatment, special procedures are required to remove solids,
etc. These are:
•
Pre-treating water with a two bed cation / anion or reverse osmosis system
•
Water treatment typical by a mixed bed deionizer.
Water Price Calculation
The following water price calculations are based on data given in reference
58. Water price is a function of several factors, ranging from daily fixed costs to
variable costs dependant from the amount of water processed. The following is a
list of these costs:
1. purchase price of the plant (amortization, Tax 8%)
2. plant maintenance
3. labor costs
4. electricity
5. raw water
6. chemicals
7. disposal
8. transport to A/C
The focus of the following calculations is to determine of the price of
conditioned water per gallon. The cost data given in (58) are for the year 1989, so a
cost price increase from 1989 to 2003 has to be recognized. Using the US
consumer price index, a 47.1% price increase is figured to have occurred in the
past 14 years. Using the maximum daily water requirement of 33,036 gallons for
Sea-Tac airport, the following water condition and cost estimations are made:
1.
The purchase price of the plant is given in (58) as $357,000 for a 20
Gallon Per Minute (GPM) output. According to the Sea-Tac example, an
average 22.94 GPM are required, which is 14.7% more than the
$357,000 plant example. As the water demand may vary extremely
NASA/CR—2004-212957
63
during the peak hours, a large storage tank will be needed to ensure
water availability during the entire day. Figuring in price inflation (47.1%)
and increased water output requirement (14.7%), the estimated cost of
the water treatment plant is:
Purchase Price = $ 357,000 * 1.147 * 1.471 = $ 602,343
Sales tax WA (8%) = $ 48,187
Unit Price = $ 650,531
Amortization of the plant is assumed to run over 15 years. On a
simple straight-line basis, the capital cost per day is:
Amortization costs =
2.
Plant Maintenance is given as 10% per year of the purchase price, which
can be easily transferred to daily fixed costs..
Daily maintenance =
3.
$650,531
= $ 118.82 per day
15 yrs ⋅ 365days
$650,531 ⋅ 0.1
= $ 178.23 per day
365
Labor is required to maintain water quality and ensure standards. Labor
costs are based on 4 hours manned supervision per day, labor cost is
based on $20.00 per hour (including benefits) in 1989. Including the price
index increase of 47.1%, labor costs are $20 ⋅ 4hrs ⋅ 1.471 = $ 117.68 per
day.
Fixed costs from above (1. to 3.) can be summarized as $414.73 per day.
The following calculations are based on an average daily water production
rate and will vary depending on the actual amount of water produced.
4.
Electricity usage is given as 2,060 kWh / day for a 20 GPM production
rate. Considering a technology improvement in the electrical motors,
pumps and overall system efficiency has probably occurred in the last 14
years, an improvement of 20% is assumed in the calculations. Electricity
costs are estimated to be $0.05/kW. Thus, the required electrical energy
is estimated to be:
Total kWh =
5.
0.05$ / kW ⋅ 2,060kWh ⋅1.147 ⋅ 0.8
= $ 0.00286 per gallon
33,036 gal / day
Raw water price is given as $ 1.18/1000 gallons. One third of the raw
water remains as wastewater and has to be disposed. Only two thirds are
usable portions of the original amount. Considering a water output of
33,036 gallons/day, 49,554 gallons of raw water are required in the
beginning of the process which results in a total price of:
$1.77/1,000 gallons treated water (output) or $ 58.47/day ($
0.00177/gallon).
NASA/CR—2004-212957
64
6.
Chemical usage is divided in H2SO4, HCl and NaOH. The Chemicals are
used for every regeneration of resin packing. The required amount is 60
lb of 30% HCl and 33.5 lb of 100% NaOH. H2SO4 is required in a 5
lb./day volume for resin regeneration in deionizer. These numbers apply
for a 20 GPM water output plant and have to be adjusted to the 22.94
GPM output.
This is 60.82 lb HCl, 38.42 lb of NaOH and H2SO4 of 5.74 lb./day.
The prices of each chemical are given on a 1989 level as:
•
H2SO4 = $ 0.22/lb.
•
HCl = $ 0.08/lb.
•
NaOH = $ 0.22/lb.
According to a 47.1% price index increase from 1989 to 2003 the
updated prices are:
•
H2SO4 = $ 0.324/lb.
•
HCl = $ 0.118/lb.
•
NaOH = $ 0.324/lb.
Applying the updated daily usages, the following daily costs occur:
•
H2SO4 = $ 0.324/lb * 5.74 lb = $ 1.860/day.
•
HCl = $ 0.118/lb * 60.82 lb = $ 7.18/day.
•
NaOH = $ 0.324/lb * 38.42 lb = $ 12.45/day.
As costs per gallon are used in the report, the sum of all three
chemicals is $21.49 per day divided by the daily volume of 33,036 gallons
of water produces the costs per gallon of:
Chemicals per gallon = $ 0.0006505 per gallon water produced
7.
Brine Hauling and disposal is necessary as the plant produces 1 gallon of
wastewater for each 2 gallons of cleaned water. Thus, the wastewater is
16,518 gallons per day. Assuming an 8,000 gallon tank capacity of a
truck would make a wastewater disposal trip twice per day. The costs per
mile of the truck are assumed to be $ 0.45 per mile in 1989, thus $0.662
in 2003 (47.1% increase). Depending on the location, a 200 mile trip is
taken as average, producing costs of $ 0.008015 per gallon. Alternately,
disposal of the brine into an existing airport industrial waste water
treatment plant would generate a cost. For this study, it will be assumed
this cost would be equivalent to the aforementioned trucking costs.
NASA/CR—2004-212957
65
The fixed cost data provided in numbers 1 to 3 above have to be broken down
into costs per gallon of conditioned water. To determine this cost, the fixed costs
are divided by the daily water production of 33,036 gallons for the example Sea-Tac
airport:
1. Installation Costs:
$ 118.82 /day = $ 0.003597 /Gal,
2. Maintenance:
$ 178.23 /day = $ 0.005395 /Gal,
3. Labor:
$ 117.68 /day = $ 0.003562 /Gal.
∑ SUM =
$ 0.01255 /Gal
The variable costs per gallon can be calculated by adding numbers 4 to 7.
4. Electricity:
$ 0.00286 /Gal,
5. Raw Water:
$ 0.00177 /Gal,
6. Chemicals:
$ 0.0006505 /Gal,
7. Disposal:
$ 0.008015 /Gal,
∑ SUM =
$ 0.0133 /Gal
Adding the results of 1-3 and 4-7 leads to overall costs of $ 0.026 per gallon.
Figure B-4 shows the cost breakdown as well as the $12.28 delivery cost to the
airplane.
Costs per Gallon (33036 Gal/day)
$ p.Gal
$0.05
$0.04
$0.04
$0.03
$0.03
$0.02
$0.02
$0.01
$0.01
$0.00
Installation Maintenance
Plant
Labor
Electricity Raw Water Chemicals
Figure B-4. Total Conditioned Water Cost
NASA/CR—2004-212957
66
Disposal
Transport
APPENDIX C. STUDY FEEDBACK
The draft report was issued to engineering, scientific and management staff at
the Boeing company as well as GE, Pratt & Whitney and Rolls-Royce engine
company staff. It was also issued to a company that specializes in providing water
injection support systems for industrial engines. The following questions and
comments were collected and answered (where feasible.)
Boeing Comments:
Comment: My initial impression on the SFC benefit for inlet misting was quite
skeptical but it is reasonable based on simple fuel flow corrections using a lower
"theta" value to account for evaporative cooling. The theta effect can also be used
to imagine there will be a significant rematch if the core is inter-cooled but the fan is
not. The report talks briefly about engine operability based on a thirty year old
paper from P&W but the rematch effect may add a "new" dimension to this
problem. For example, N2 rotor speed (along with SFC) would be reduced by
nearly 3% for the 100 degF sample case where misting lowers core inlet conditions
to 69 degF, but N1 would remain essentially unchanged.
Comment: Although there would be a high initial airport non-recurring cost, it
would probably make sense to install a dedicated purified water line at each gate at
the airport. Since the service cost is the highest part of providing the aircraft with
purified water, this cost could be eliminated by having the ground service personnel
refill the potable and purified water tanks at the same time.
Comment: The water injection system maintenance costs you listed in figure
4.27 should probably be reduced, or maybe even eliminated. We can do a lot
better today than those old TWA numbers you quoted.
(author’s comments: from the above two comments … by eliminating the
water delivery charge and reducing the system maintenance costs 80% would cut
the operator costs of the system about in half [$20.39/takeoff] and improve the
cost/benefit ratio [$832/ton] of the NOx reduced for this study airplane. )
Comment: An 80 mile decrease in range is not insignificant and needs to be
addressed. A dollar value on this performance penalty number needs to be figured
somehow and included in the cost/benefit analysis.
Comment: Adding more components to current installations will be a
challenge. There isn’t a lot of space available on current engines to add more
components without impacting repair times in a negative way. The risk is that this
new system will potentially block access to other systems on the engine/strut. If the
installation blocks another system you suffer the penalty of having to remove water
injection system components (lengthening repair times and potentially inducing
damage) when the blocked system has a failure requiring maintenance. The
converse is also true, if the installation buries components of the water injection
system, when it fails you may have to remove components not part of the water
injection system (increasing maintenance time). Depending on which components
NASA/CR—2004-212957
67
have to be removed and reinstalled, there may be an engine run required to verify
proper system operation. Again, increased maintenance time and cost (fuel,
additional engine wear).
Comment: A 400+ degree decrease in T4 is amazing. We might even see
something like a doubling of turbine life with that kind of reduction.
Q: How many breaks/connections in the system are there? Every one of them
is a potential site for leakage that negatively affects system operation and will
require maintenance. There is the added problem of potential corrosion at each
one of those sites.
A: The number of connections will be minimized where possible. One
operator has reported corrosion problems in their water injected 747s. As the wing
tanks have bladders, perhaps it is feasible that this corrosion could be coming from
the water lines and connections.
Q: What method(s) will be used to determine if there is a system failure? Is it
integrated with engine monitoring systems? Is it separate? Are there no
diagnostics at all?
A: There is a water tank level gage and a water flow gage that will help with
diagnostics. However, if a water line ruptured or leaked inside the wing there may
be no way to determine this. The water would freeze and could cause problems
with another system. A failure detection system needs to be addressed.
Q: If the system has a failure can the airplane continue to fly without any
needed maintenance action required until a time is found to repair the system?
A: The system is designed to be operated at the discretion of the pilot and
operator. That is, it doesn’t need to be used for added thrust during takeoff. It is
only for NOx control. However, if the system would fail with water still in the tanks,
it must be serviced to remove the water. Existing water injection systems have not
been affected by having a small amount of water remain in the tanks and pumps
that froze up.
Q: What are the field length impacts of provisioning for the system not running
and you have to complete the takeoff with 2,500 lbs of extra water? What if the
system failed to drain? What are the operating cost impacts of these failures?
A: The study airplane did not take advantage of the potential additional thrust
available and included 2,865 of added weight (2,505 for water and 360 for system),
displacing that same amount of fuel weight, so there would be no impact on field
length if the system failed. If it experienced a double failure (didn’t drain) then the
airplane would have to return to the airport. Double failure scenarios are generally
not included in cost calculations.
NASA/CR—2004-212957
68
Q: What is the impact of the small amount of extra thrust (or does the control
throttle back to hold thrust constant?) Would this be allowed?
A: The study airplane kept the same thrust level when using water misting, so
the engines would be throttled back slightly. Engine control scenarios need to be
addressed by the engine companies.
Q: Although we think the range penalty is small, what happens to
range/payload when you are trying to takeoff out of Denver on a hot day on a flight
to JFK? Would it be impacted by having to plan for the extra water weight?
A: When the airplane is filled to 100% capacity and is range limited, perhaps
the air quality authorities would allow takeoff without water injection for these
infrequent episodes? Ironically, it is these instances where the added thrust from
the water injection could be most used, but was not considered due to safety
concerns if the system failed.
Comment: - You're adding a lot of water to the combustor and water is not
inert at those temperatures. It will have some effects on the heat distribution and
chemistry which may influence the other emissions. I would be concerned about
how the water injection affects the fuel droplet dispersion and evaporation and inturn how that affects mixing/soot generation and combustor efficiency (i.e., CO and
hydrocarbons) I suspect that water injection may increase hydrocarbon and soot
emissions if it interferes with fuel droplet evaporation and fuel/air mixing. You're
going to change the heat distribution in that region by injecting water and you don't
seem to be addressing or even commenting on that in any way
Comment: You assume that the empirical T3/P3 relationships will work even
though you have made major changes to the heat capacity and composition in the
combustor. This makes me uncomfortable without some kind of analysis to support
this assumption.
Comment: Note that while water injection has been used in stationary power
plants, these are systems designed for single-point operation. For aircraft the
combustor has to work over a much wider dynamic range. Water injection may be
trickier, particularly since you want to inject water at the power settings that are
most important for takeoff and safety of flight. What happens if the water system
dumps too much water into the combustor? Fire goes out at the wrong time?
Comment: Reducing NOx at the expense of these other emissions [HC, CO,
Soot], may or may not be a good trade.
Comment: In figure 4.25, you contrast CO generation for 2 engines (LM2500
and LM6000). Other than telling us CO may be a big problem, what does this
figure tell us about how an aircraft engine would behave?
NASA/CR—2004-212957
69
From CH2M Hill, Bellevue, WA (H20 conditioning support company)
Comment: Water conditioning costs seem high. Without including
transportation costs, Boeing comes up with 0.026 $/gallons for conditioned water.
Portable units on the market right now cost less than $0.02 $/gallon, which would
be for the worst case scenario. For fixed base units, the costs would be around
$0.01 $/gallon. However, this does not include brine disposal costs that would
increase the cost.
Comment: Capital acquisition costs also appear high. Boeing quotes a
$650,531 capital acquisition cost. A comparable performance unit today costs
about $300,000. New Reverse Osmosis (RO) technology has dramatically brought
down the cost of conditioning water from the days when the GRI report (that Boeing
used as a basis) was written.
Comment: Water delivery costs, which make up the biggest portion of the
water injection cost, might be brought down by installing piping throughout the
airport. At the $12.28 per service cost that Boeing figured, it probably wouldn’t take
long to pay off such a piping system at the airport. One needs to be careful to use
either stainless steel piping (expensive) or plastic (affordable) to transport the
conditioned water as other pipe materials can corrode very quickly when carrying
this pure water. Another option might be to „piggy back“ onto the existing potable
water delivery truck.
Comment: Maintenance time is too high. Today’s water conditioning plants
require a lot less maintenance than earlier generation plants. The 4 hours per day
that Boeing quotes could probably be reduced to ½ to 1 hour per day.
Comment: For any next round of studies, it would probably make sense to
look at several different sizes and types of airports as well as put more time into
optimizing the water conditioning costs. This would make the study more realistic
and would also probably result in lower water costs used to evaluate the overall
cost of the technology.
Comments from GE Aero Engine Group:
The Boeing Draft Report on Commercial Aircraft Water Misting and Injection is
an interesting and reasonably comprehensive initial evaluation of water injection.
As a general comment, the cost/benefit numbers stated in the draft report may be
overly optimistic. As discussed in our comments below, factors such as the
potential usable range of water-to-fuel ratio, airplane size, effective cost of lost
payload/range capability and landing charges are not fully considered in the initial
cost/benefit analysis.
Water injection may be effective for low altitude operations, but it cannot be
expected to address cruise emissions. In the case of the advanced 777 studied in
the draft report, ~2500 Lb of water was required to reduce NOx by 46.5% during the
first few minutes of each flight. Since the water would be expended at the start of
flight, impact on mission fuel consumption appears to be acceptable in most cases,
but it would be impractical to carry enough water for cruise. Therefore, low
NASA/CR—2004-212957
70
emissions combustors will still be needed to reduce NOx at cruise. Water injection
and application of low emissions combustor technology could be complementary.
Water injection could be used at takeoff, where the combustor inlet conditions
make it most difficult to apply low emissions technologies, and the combustor could
be optimized to reduce NOx at cruise.
In the case of an aircraft on a maximum range mission, it might be necessary
to depart without water injection in order to safely meet the mission range
requirement.
In the past, emissions reduction has been considered to be primarily an
engine issue. With water injection, the responsibility for emissions reduction is
shared with the airframe (water tanks and pumps) the airline (operation and
maintenance of the system) and possibly the airport (treated water infrastructure).
A complete evaluation of water injection for reduction of NOx emissions must
consider engine technology, airframe requirements, servicing operations and
infrastructure. The draft report covers all of these aspects for one aircraft
application, the advanced 777. While recognizing that there could be major issues
with respect to airframe, servicing and infrastructure, GEAE comments will focus on
engine technology, our primary area of expertise.
Safety – Any approach to reduce emissions must be proven to be safe.
Intuitively, spraying water into a fire suggests the possibility that the fire will be
extinguished. The added complexity and servicing requirements of a water
injection system also provide new opportunities for system reliability issues that
could affect critical takeoff operations. The draft report describes previous
experience with water injection, so safety issues have presumably been addressed.
GEAE does not have experience with water injection in commercial engines, and
we believe that a thorough system failure analysis must be completed before we
could agree that water injection is an acceptable alternative for aircraft emissions
reduction. Some risks that come to mind include:
- Water pump failure - Is there another way to empty the tanks? How much would
range be reduced? Would water freezing in tank cause damage?
- FOD from water injectors in combustor or compressor
- Excessive water injection rate - Can it cause thrust loss?
- System leakage or backflow into water tank – Water injectors are in
communication with fuel and combustor inlet air at >1200F and >600psi
- Icing would be a particular concern with compressor injection – If alcohol was
used to prevent icing, most would bypass the combustor and be emitted as HC.
NOx Reduction May Be Overstated – The NOx reduction values calculated
in the section on performance impacts (Table 4.3) and carried over into the
discussion of NOx reduction cost (Table 4.4) is 49.2 Lb. NOx per takeoff. The
ICAO data bank indicates total NOx emissions for two GE90-85 engines to be 107
lb for the ICAO LTO Cycle. Of the total, about 81 Lb. is produced at takeoff and
climb conditions where water is injected. Based on that figure, at 46.5% NOx
reduction, the total amount of NOx reduced would be about 38 Lb. Additionally, the
NASA/CR—2004-212957
71
ICAO LTO cycle is somewhat conservative with respect to NOx emissions because
a modern twin-engine aircraft will typically operate at derated thrust, and will climb
faster than assumed by the LTO cycle. This will probably reduce the typical inservice benefit by another 25%, to about 29 Lb. Assuming the costs are correct,
this reduction in NOx benefit would increase the cost from $1,663 to $2,800 per ton
of NOx reduced.
NOx Reductions Relative to Cost Based on 777 Aircraft May Not Be
Typical of The Fleet - The engines that power the Boeing 777 aircraft operate at
much higher pressure and temperature most than other engines in service. The
result is that NOx emissions are high relative to other engines. For example, the
NOx emissions index for the GE90-85 at the ICAO takeoff operating condition is
47.28g/kg (Figure 4.20), nearly twice as high as the value of 25.30 g/kg for a
CFM56-7B24 that is used on a 737. Since the NOx reduction benefit is roughly
proportional to the NOx EI, the benefits might be reduced by nearly 50% in the 737
(thereby doubling the Cost/Benefit ratio). To provide a balanced perspective, cost
and benefit numbers should be estimated for the 737. This is an important
consideration because narrow body aircraft are major contributors to NOx
emissions (resulting from the large number of narrow body aircraft and the
tendency to conduct more operations per aircraft per day). On the cost side, cost
per unit of NOx emissions reduced will tend to be increased for smaller aircraft
because some costs (e.g. water servicing cost) are not dependent on aircraft size.
Status of Compressor Water Injection Technology – Compressor water
injection for industrial engine performance enhancement has only been applied
over the past few years. Therefore, there are still some issues that might still have
to be addressed (e.g. water injector life, compressor erosion) with respect to longterm operation.
Compressor Water Injection Rate – In industrial applications, the draft report
indicates that there is experience with injection rates between 0.5 and 0.87% of
core airflow on 90F days. The 2.2% rate assumed for aircraft applications in this
study would appear to be well beyond industrial engine experience. Potential
effects on compressor erosion, NOx emission reduction effectiveness, effect on
other emissions, icing and engine stability would have to be considered.
NASA/CR—2004-212957
72
Status of Combustor Water Injection Technology – Combustor water
injection for NOx reduction has been in use in ground based industrial engines for
over 15 years.
Combustor Durability - Initial combustor durability issues due to water erosion
have been addressed, and continuous operation with water injection for over
20,000 hours has been achieved in some cases. Therefore, based on total
injection time, reduced durability with aircraft water injection would not be expected
to be a major issue; however, there could still be issues with cyclic (LCF) life in
aircraft applications.
Range of Operation - Water injection has been used in industrial engines
derived from CF6-6, -50, and –80C aircraft engine designs. It has been used
successfully over a moderate range of engine pressure ratio and turbine inlet
temperature. Effectiveness in very high pressure and temperature engines such as
those used on the 777 would still have to be demonstrated. NOx reduction is not
expected to be a barrier, but tradeoffs with CO might be significant in such
applications.
Turbine Durability - Reduced turbine inlet temperature with water injection
could significantly improve hot section durability.
Combustor Water Injection Rate – In industrial applications, combustor
water injection has proven effective in reducing NOx emissions by up to 90% at
steady state operating conditions with water-to-fuel ratios of about 1.0. In some
applications there is a tradeoff with CO emissions at this level. If the aviation
industry took on the cost and complexity of water injection, there would be pressure
to maximize the benefit by using a water-to-fuel ratio of 1.0. However, in order to
provide margins for CO emissions and stability during aircraft operations
(particularly rapid engine transients that might be required during takeoff and climb),
it would be more prudent to consider limiting water-to-fuel ratios of about 0.7 in
aircraft applications. This would still provide NOx reduction of up to 75%. As
indicated in the table below, this would also limit the water weight to about 1500
Lb., and constrain the tendency to increase CO emissions.
NASA/CR—2004-212957
73
Compressor vs. Combustor Injection – The draft report seems to favor
compressor injection (misting) based on improved engine performance and
potential performance retention. However, combustor injection has advantages in
that it can achieve greater NOx reduction, requires less water, and is based on
technology that has been proven in long-term industrial service. The comparisons
in Tables 4.3 and 4.4 of the draft report do not consider that with the smaller
amount of water needed for combustor injection, the impact on payload/range
capability is also smaller. Specifically, in Table 4.3 we estimate that the range loss
for combustor injection (MTOW limited) should be about 42 nmi, compared to ~80
nmi for LPC injection.
Another way to compare the options is to assume that the potential of combustor
injection to further reduce NOx (75% reduction) is used. A rough comparison
between combustor injection for 75% NOx reduction to LP compressor injection for
46.5 % reduction is shown below.
Percent NOx Reduction Base
0
50
75
85
Water to Fuel Ratio
0.00
0.00
0.35
0.70
1.00
Fuel Burn to 3000ft., Lb 2000
2000
2017.5
2035
2050
Weight of Additional
Fuel
0
0
17.5
35
50
Weight of Water Tank
and Pumps, Lb
0
360
360
360
360
Weight of Water
0.0
0.0
706.1
1424.5
2050.0
Total Water Weight
Adder (lost payload), Lb 0.0
360.0
1083.6
1819.5
2460.0
Range Reduction, nmi
0.0
10.4
31.2
52.5
70.9
NOx Reduced, Lb
0.0
0.0
40.5
60.8
68.9
CO Increased, Lb
0
0
0
1.8
6
CO Increased, %
0.0%
0.0%
0.0%
6.3%
20.9%
NASA/CR—2004-212957
74
Percent NOx Reduction
during Takeoff and Climb
Water to Fuel Ratio
Fuel Burn to 3000ft., Lb
Weights
Weight of Additonal Fuel
(Relatrive to Baseline)
Weight of Water Tank &
Pumps, Lb
Weight of Water
Total Water System Weight
Adder (lost payload), Lb
Range
Range Reduction, nmi
Emissions
LTO NOx Emitted, Lb
NOx Reduced, Lb
CO Increased, %
Combustor
75
LP Compressor
47
0.70
2035
1.02
1910
35
-90
360
360
1425
1820
1940 *
2210
52
64
46
61
6%
69
38. *
0%
*Values of NOx emission reduction and weight of water from draft report reduced consistent
with the total NOx reported on the ICAO data sheet.
With combustor injection, NOx emissions are reduced by 61 Lb, compared to
38 Lb with compressor injection. Even with this further NOx reduction, less water is
needed with combustor injection, so the impact on payload is 290 Lb less than with
LPC injection. Combustor injection is also based on more mature technology.
There would be a slight increase in cost to account for 125Lb (~15 gal.) fuel use
with combustor injection, and CO would be increased very slightly. Taking all of
these effects into consideration the cost per ton of NOx reduced would likely be
about 30% less with combustor injection than with compressor injection.
Increased Thrust with Water Injection - If water injection is used to reduce
NOx emissions, there will be a temptation to use it for thrust augmentation. If this
capability is used, water availability and water injection reliability become dispatch
critical, and may have a greater impact on flight safety. Therefore, we agree with
your view that studies should not count on water for thrust augmentation.
Costs – The simple cost estimates in the report should to be examined by the
operators to confirm that all aspects of cost (e.g. spares, delays, lost payload/range
capability, cost of money) have been considered. More detailed studies can also
be expected to reveal unanticipated cost items. As an example, water injection to
control NOx during typical reduced thrust operations will probably require more
precise control of water flow than has been used for thrust augmentation, so control
system costs will likely be higher than estimated in the draft report. Sensitivity of
costs to assumptions such as aircraft utilization needs to be estimated. For
example, GEAE would base cost estimates on ~640 trips/year, rather than the 475
trips/year assumed in Table 4.4. Utilization will vary depending on the operator, so
it might be more useful to show a range of costs corresponding to a probable range
of input assumptions.
NASA/CR—2004-212957
75
Lost revenues due to reduced payload could far exceed all other costs associated
with water injection. For example, on flights where range is critical, it would be
necessary to displace cargo with water for NOx control. For ~50% NOx reduction,
we estimate that this could result in lost revenue of $720 (for combustor injection) to
$1285 (for compressor injection) for a 6000 nmi flight. This would not occur on
every flight, but the impact should be considered.
One alternative to avoid lost revenue would be to increase MTOW to recover
payload/range capability. This was studied in recent ICAO FESG analysis of
proposed new NOx standards, where range was potentially reduced due to
increased fuel consumption. FESG assumed that airlines would have to buy
additional MTOW as necessary to maintain aircraft payload/range capability at a
cost of $217 per additional pound of MTOW (based on public price data). We
estimate that it would require about 3000 lb additional MOTW with LPC injection to
account for the weight of water, tanks, pumps and additional fuel. Based on FESG
numbers, incremental airplane purchase price could increase by more than
$600,000. We believe that the actual costs would be somewhat lower, ranging
from $110,000 (combustor injection) to $200,000 (LPC injection) for ~50% NOx
reduction. Using the assumption that capital cost is spread over 11,875 departures
(25 year life, 475 trips per year), this could add up to $10 to $50 per departure.
Additionally, FESG estimated a significant increase in operating cost due to
increased landing fees that are based on MTOW. FESG used an international
average value of $4.50 per 1000 Lb of MTOW. At that rate, the 2500 Lb increase
in MTOW would add another $11 per departure. Independent GE estimates of
incremental landing fees range from slightly more than $5 (combustor injection) to
slightly more than $9 (LPC injection) for 50% NOx reduction.
Based on the above discussion, the cost per departure due to increased MTOW is
potentially larger than the total delta cost of $40.92 to $52.90 estimated in Tables
4.4 and 4.5 of the draft report. This could more than double the cost per ton of NOx
reduced.
Comments from Pratt & Whitney:
1) This is a very good and timely feasibility study. It shows that there is a
significant potential benefit of NOx reduction by water injection. It also shows that
from an airplane/engine system point of view, carrying water is feasible and may be
cost effective. However, while it may be argued by some that water misting is
cheaper than expensive combustors, past history with water injection has proven
that adding water is more expensive than not adding water. There may be
efficiency gains and NOx reductions, but the impact of adding water to short, high
intensity combustors must be ascertained.
2) As the study points out, there are several options for adding water with
varying and adverse system impacts. The impact on compressor surge margin has
been mentioned, but no serious study has been conducted to quantify it. This may
be a show stopper for introducing water before the HPC.
NASA/CR—2004-212957
76
3) Where and in what quantity the water is added will also depend on the
combustor design and the system will have to be optimized taking that into account.
The history of water injection in aircraft engines is based on rich front-end
combustors. The study implies, although not explicitly states, that future
combustors will be lean front-end. We do not have any history with those
combustors in aircraft engines. It is also not a forgone conclusion that all future
combustors will be lean front-end.
4) Pratt & Whitney has developed expertise in NOx reduction with water
injection in aero-derivative engines for industrial application with rich front-end
combustors. P&W has demonstrated the capability to reduce NOx by a factor of
10X in industrial applications.
5) If Boeing proceeds with further conceptual studies to demonstrate the
viability of water injection to reduce NOx in aero engines, Pratt & Whitney would be
willing to work with Boeing and a partnership of other engine companies to design
and develop an aircraft/engine system.
6) It appears that business case development costs are based on lean staged
combustors. The development costs associated with RQL technology must also be
assessed.
7) Impact on CO emissions of rich and lean concepts must be assessed.
8) Water injection will increase engine thrust, hence the potential benefit to
fuel burn (CO2) that may occur during take-off and climb needs to be assessed.
Overall, the final report of the Boeing study on water injection is well done as
far as performance, emissions and system design are concerned. The concern is
that the operability issues with regard to water injection are treated in only a
minimal fashion. Two major areas of concern are not adequately addressed. First,
that of water droplet size and second, that of HPC stability changes due to water
injection. These two areas are major drivers in the success or failure of the water
injection system and should be further addressed in the report, or the overall
conclusions are too easily accepted as easy opportunities [author’s note -- further
information on HPC operability effects are contained in P&W’s operability
memorandum WTC-04001 from William T. Cousins dated 1/16/04]
NASA/CR—2004-212957
77
REFERENCES
1
Emissions Related Landing Charges Investigation Group, Paris France, 6
October, 2003
2
Jones, R.E. and Grobman, J., Design and Evaluation of Combustors for
Reducing Aircraft Engine Pollution, Paper 31, Atmospheric Pollution by Aircraft
Engines, AGARD CP-125, Advisory Group for Aerospace Research &
Development, 1973
3
Ingebo, R.D. and Norgren, C.T., Effect of Primary-Zone Water Injection on
Pollutants form a Combustor Burning Liquid ASTM A-1 and Vaporized Propane
Fuels, NASA Technical Report TN D-7293, May 1973
4
Marchionna, N.R., Diehl, L.A., and Trout, A.M., The Effect of Water Injection
on Nitric Oxide Emissions of a Gas Turbine Combustor Burning ASTM Jet-A Fuel,
NASA Technical Memorandum X-2958, 12/1973
5
Hilt, M.B. and Johnson, R.H., Nitric Oxide Abatement in Heavy Duty Gas
Turbine Combustors by Means of Aerodynamics and Water Injection. ASME Paper
72-GT-53, 1972
6
Klapatch, R.D., and Koblish, T.R., Nitric Oxide Control with Water Injection in
Gas Turbines, ASME Paper 71-WA/GT-9, 1971
7
Balepin, V., Ossello, C, and Snyder, C., NOx Emission Reduction in
Commercial Jets through Water Injection, AIAA Paper 2002-3623, 7/7/2002
8
Controlling Airport-Related Air Pollution, Northeast States for Coordinated
Air Use Management and Center for Clean Air Policy, June 2003
9
Aviation and the Environment, Strategic Framework Needed to Address
Challenges Posed by Aircraft Emissions, GAO Report to the chairman,
Subcommittee on Aviation, Committee on Transportation and Infrastructure, House
of Representatives, GAO-03-252, February 2003.
10
Northeast States for Coordinated Air Use Management and Center for
Clean Air Policy, Controlling Airport-Related Air Pollution, June 2003
11
Blazowski, W.S and Henderson, R.E., Aircraft Exhaust Pollution and its
effect on the U.S. Air Force, AFAPL-TR-74-64, AD 783828, Air Force Aero
Propulsion Laboratory, Wright-Patterson Air Force Base, Ohio 1974
12
Ding, J.M.W. et. al, External costs of aviation, CE, Solutions for
environment, economy, and technology, www.ce.nl, Delft, Netherlands, February
2002
13
Keenan, J.H., Kaye, J., Gas Tables, Thermodynamic Properties of Air
Products of Combustion and Component Gases, John Wiley & Sons, New York,
1948
NASA/CR—2004-212957
78
14
Lefevbre, A.H. Gas Turbine Combustion, Taylor & Francis, 1983
15
J. Tilston, et. al., Future Engine Cycle Prediction and Emissions Study,
QinetiQ Final Report, Contract G4RD-CT-2000-00383, August 2003
16
Bahr, D.W., Technology for the Reduction of Aircraft Turbine engine
Exhaust Emissions, Paper 29, Atmospheric Pollution by Aircraft Engines, AGARD
CP-125, Advisory Group for Aerospace Research & Development, 1973
17
Daggett, D.L., et. al. Ultra Efficient Engine Technology Systems Integration
and Environmental Assessment, NASA CR 2002-211754. 2002.
18
International Civil Aviation Organization (ICAO), International Standards and
Recommended Practices, Environmental Protection, Annex 16 to the Convention
on International Civil Aviation, Volume II, Aircraft Engine Emissions, 2nd Edition,
Montreal, 1993.
19
Lefebvre, A. H. Gas Turbine Combustion, Taylor & Francis, 1999.
20
Personal conversation between author and Arturo Benito of Iberia airlines,
2000
21
Koch, H. and Felix, P., Exhaust Gas Emissions of Brown Boveri Gas
Turbines, Brown Boveri Rev., Vol 64, no. 1, pp 27-33, 1977
22
Zucrow, M.J. Jet Propulsion and Gas Turbines, John Wiley & Sons, New
York, 1948
23
The Jet Engine, Rolls-Royce plc., Derby, England, ISBN 0 902121 2 35,
1986
24
Mee III, T.R., Reduction of Gas Turbine NOx Emissions by Inlet Air
Fogging, MEE Industries, Monrovia, CA, 12/30/1999
25
The Aircraft Gas Turbine Engine and its Operation, Pratt & Whitney Aircraft,
E. Hartford, CT, May 1974.
26
Klein, V.J., “Engine Water Injection System – Description, Performance and
Failure Analysis”, Boeing Document Number D6-30549, 11/12/1971
27
Wilcox, R. S; Cost Effectiveness of Large Aircraft Engine Emissions
Controls, EPA Document 61-11-EPA460/3-80-009, December 1979.
28
“Water Injection System Improvements – 747”, Boeing Airlines Magazine,
Page 10-11, October 1971
29
Boeing Airliner Magazine, Page 22, Oct-Dec 1990
30
Radloff, P., Personal discussion regarding McDonnel Douglass’ experience
with water injection on DC-10-40 aircraft during flight test, October 21, 2003
31
Water Injection Study for GE90-115B, GE/Boeing Meeting on August 22,
32
Water Injection Q&A, P&W/Boeing Meeting on 3/13/03
2002
NASA/CR—2004-212957
79
33
Urbach, H.B., et. al. Water Injection into Navy Gas Turbine Combustors to
Reduce NOx Emissions, ASME Paper 98-GT-298, 6/2/1998
34
U.S. Department of Labor, Bureau of Labor Statistics,
http://www.bls.gov/cpi/#tables
35
Henderson, R.E. and Blazowski, W.S., Aircraft Gas Turbine Pollutant
Limitations Oriented toward Minimum Effect on Engine Performance, Paper 33,
Atmospheric Pollution by Aircraft Engines, AGARD CP-125, Advisory Group for
Aerospace Research & Development, 1973
36
Personal discussion with Nigel Bond, Rolls-Royce plc., November 2003
37
Part 87, Title 40, Code of Federal Regulations, Environmental Protection
Agency; Control of Air Pollution from Aircraft and Aircraft Engines, Emissions
Standards and Test Procedures for Aircraft, Federal Register, July 17 1973.
38
Snodgress Jr., J.C., Comments on EPA Emissions Standards for Aircraft
Turbine Engines, AIAA Position Paper on Water Injection, AIAA Washington DC,
January 23, 1976
39
Ross, P.A., Ground Rules and Basis for Preliminary Design and Costing of
Water Injection Systems for Emissions Control in the 737, 727 and 747 Airplanes,
Boeing Memo B-8442-2441, Seattle, WA. October 10, 1975
40
Daggett, D.L., Eelman, S., Kristiansson, G.; Fuel Cell APU for Commercial
Aircraft, AIAA-2003-2660, 2003
41
Hendricks, R.C., Shouse, D & Roquemore, M; Water Injected
Turbomachinery, The 10th International Symposium on Transport Phenomena and
Dynamics of Rotating Machinery, 2004.
42
Hung, W.S.Y., Accurate Method of Predicting the Effect of Humidity or
Injected Water on NOx Emissions from Industrial Gas Turbines, ASME report 74WA/GT-6, November 1974.
43
Daggett, D.L., et. al; Ultra-Efficient Engine Diameter Study, NASA
Contractor report # NASA/CR-2003-C03-003, January 2003
44
“Intercooler Boosts LM6000 output 9%”, Gas Turbine World, Page 44,
November-December 1998.
45
Hines, W.R., Gas Turbine Engine with Water Injection, U.S. Patent
6,012,279, 1/11/2000
46
Wilkes, C. Control of NOx Emissions from Industrial Gas Turbine
Combustion Systems”, Air and Waste Management Association, 82nd Annual
Meetings and Exhibition, Anaheim, CA, June 25-30, 1989.
47
Personal phone call with Carl Shook of GE Industrial Engine Systems.
October 8, 2003
48
International Civil Aviation Organization (ICAO), ICAO Engine Exhaust
Emissions Data Bank, 1st Internet Edition,
NASA/CR—2004-212957
80
49
Koch, H. and Felix, P., Exhaust Gas Emissions of Brown Boveri Gas
Turbines, Brown Boveri Rev., Vol 64, no. 1, pp 27-33, 1977
50
Shaw, H., The Effect of Water on Nitric Oxide Production in Gas Turbine
Combustors, ASME Paper 75-GT-70, 1975
51
Bahr, D. W.; Lyon, T.F.; NOx Abatement via Water Injection in Aircraft
Derivative Turbine Engines, ASME Paper 84-GT-103, 1984.
52
Flite Line Equipment corporation, Model FL-1000 water service truck,
Miami, FL. 305-626-0004, www.fliteline usa.com
53
Mutual Assistance Ground Service Agreement, Industry Accepted Rates,
Page 2, 2002
54
Castaldini, C., Evaluation of Water Injection Impacts for Gas Turbine NOx
Control at Compressor Stations, Gas Research Institute Paper #PB90-259979
55
http://www.portseattle.org/factstat/default.htm
56
http://www.portseattle.org/factstat/stats/air/ArptActvRpt.pdf
57
Evaluation of Water Injection Impacts for Gas Turbine NOx Control at
Compressor Stations, Topical Report 06-09/1989, Gas Research Center Institute,
Chicago, Il, P.15
NASA/CR—2004-212957
81
Form Approved
OMB No. 0704-0188
REPORT DOCUMENTATION PAGE
Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources,
gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this
collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson
Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188), Washington, DC 20503.
1. AGENCY USE ONLY (Leave blank)
2. REPORT DATE
3. REPORT TYPE AND DATES COVERED
Final Contractor Report
March 2004
4. TITLE AND SUBTITLE
5. FUNDING NUMBERS
Water Misting and Injection of Commercial Aircraft Engines to Reduce
Airport NOx
6. AUTHOR(S)
Cost Center 2250000013
GRC Purchase Order C–74711A
David L. Daggett
8. PERFORMING ORGANIZATION
REPORT NUMBER
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)
Boeing Commercial Airplane Group
P.O. Box 3707
Seattle, Washington 98124–2207
E–14397
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES)
10. SPONSORING/MONITORING
AGENCY REPORT NUMBER
National Aeronautics and Space Administration
Washington, DC 20546– 0001
NASA CR—2004-212957
C&EA–BQ130–Y04–002
11. SUPPLEMENTARY NOTES
Project Manager, Robert C. Hendricks, Research and Technology Directorate, NASA Glenn Research Center,
organization code 5000, 216–977–7507.
12b. DISTRIBUTION CODE
12a. DISTRIBUTION/AVAILABILITY STATEMENT
Unclassified - Unlimited
Subject Categories: 01, 07, and 45
Distribution: Nonstandard
Available electronically at http://gltrs.grc.nasa.gov
This publication is available from the NASA Center for AeroSpace Information, 301–621–0390.
13. ABSTRACT (Maximum 200 words)
This report provides the first high level look at system design, airplane performance, maintenance, and cost implications
of using water misting and water injection technology in aircraft engines for takeoff and climb-out NOx emissions
reduction. With an engine compressor inlet water misting rate of 2.2 percent water-to-air ratio, a 47 percent NOx reduction was calculated. Combustor water injection could achieve greater reductions of about 85 percent, but with some
performance penalties. For the water misting system on days above 59 ∞F, a fuel efficiency benefit of about 3.5 percent
would be experienced. Reductions of up to 436 ∞F in turbine inlet temperature were also estimated, which could lead to
increased hot section life. A 0.61 db noise reduction will occur. A nominal airplane weight penalty of less than 360 lb
(no water) was estimated for a 305 passenger airplane. The airplane system cost is initially estimated at $40.92 per
takeoff giving an attractive NOx emissions reduction cost/benefit ratio of about $1,663/ton.
15. NUMBER OF PAGES
14. SUBJECT TERMS
Aircraft emissions; Water injection; Misting; Turbomachines; Aircraft; Jet engines
17. SECURITY CLASSIFICATION
OF REPORT
Unclassified
NSN 7540-01-280-5500
18. SECURITY CLASSIFICATION
OF THIS PAGE
Unclassified
19. SECURITY CLASSIFICATION
OF ABSTRACT
94
16. PRICE CODE
20. LIMITATION OF ABSTRACT
Unclassified
Standard Form 298 (Rev. 2-89)
Prescribed by ANSI Std. Z39-18
298-102
Fly UP