U.S. patent application number 11/023035 was filed with the patent office on 2006-12-28 for system for detecting and compensating for aerodynamic instabilities in turbo-jet engines.
Invention is credited to John Chivers, Alan Epstein, Anthony D. Kurtz.
Application Number | 20060288703 11/023035 |
Document ID | / |
Family ID | 37565649 |
Filed Date | 2006-12-28 |
United States Patent
Application |
20060288703 |
Kind Code |
A1 |
Kurtz; Anthony D. ; et
al. |
December 28, 2006 |
SYSTEM FOR DETECTING AND COMPENSATING FOR AERODYNAMIC INSTABILITIES
IN TURBO-JET ENGINES
Abstract
The present invention relates to a system for detecting
aerodynamic instabilities in a jet turbine engine having a pressure
transducer mounted in the engine. The pressure transducer, welded
to a circuit in signal communication with a controller, is adapted
to send measured pressure readings from air in a combustion chamber
to the controller. The controller, located in spaced apart relation
from the engine, is adapted by software to detect pressure patterns
from the pressure signals generated by the transducer that are
indicative of a stall or surge. A series of fuel and air valves
located with compression and combustion chambers of the engine are
in signal communication with the controller. The controller in
response to detecting pressure signals indicating a stall or surge
is operative to signals in the valves to change the air flow, air
angle, fuel flow or speed to reduce the possibility of a stall or
surge.
Inventors: |
Kurtz; Anthony D.;
(Ridgewood, NJ) ; Chivers; John; (Derby, GB)
; Epstein; Alan; (Lexington, MA) |
Correspondence
Address: |
PLEVY & HOWARD, P.C.
P.O. BOX 226
FORT WASHINGTON
PA
19034
US
|
Family ID: |
37565649 |
Appl. No.: |
11/023035 |
Filed: |
December 23, 2004 |
Current U.S.
Class: |
60/772 ;
60/803 |
Current CPC
Class: |
F04D 27/001 20130101;
F23N 2241/20 20200101; Y02T 50/60 20130101; F02C 9/28 20130101;
F05D 2270/301 20130101; F01D 17/08 20130101; F01D 21/003 20130101;
F05D 2260/80 20130101; F23N 2225/04 20200101; G01M 15/14 20130101;
F05D 2300/50 20130101; F23N 5/16 20130101; Y02T 50/672
20130101 |
Class at
Publication: |
060/772 ;
060/803 |
International
Class: |
F02C 7/00 20060101
F02C007/00 |
Claims
1. A method for accurately predicting aerodynamic instabilities in
turbo-jet engines having at least one compression chamber
comprising the steps of: placing a plurality of pressure
transducers about a circumferential wall of the at least one
compression chamber; initially forming said pressure transducers by
fusion bonding an insulating layer of a carrier wafer to a wafer
having piezoresistive regions to create a high temperature
transducer; said placing step further including positioning each of
said pressure transducers about the circumference to collectively
measure aerodynamic excitation frequencies; measuring pressure
fluctuations in said at least one compression chamber; and
associating measured pressure fluctuations with aerodynamic
excitation frequencies associated with a aerodynamic
instability.
2. The method of claim 1 wherein said bonding step is diffusion
enhanced fusion bonding.
3. The method of claim 1 wherein said initial forming step includes
forming a resistance bridge circuit on said piezoresistive
wafer.
4. The method of claim 1 wherein said associating step includes
processing said measured fluctuations using Fourier transforms.
5. The method of claim 1 wherein said initial forming step includes
forming said insulating layer by oxidizing a surface of said
carrier wafer.
6. The method of claim 1 wherein said at least one compression
chamber includes intermediate and high compression chambers and
said placing step includes placing said transducers about the
circumference of said intermediate compression chamber.
7. The method of claim 1 wherein said at least one compression
chamber includes intermediate and high compression chambers and
said placing step includes placing said transducers about the
circumference of said high compression chamber.
8. The method of claim 1 wherein said initial forming steps
includes incorporating a vibration compensator into said pressure
transducer.
9. The method of claim 1 used in a compensation process for a
turbo-fan engine further including a combustion chamber and
actuators to adjust air flow in said at least one compression
chamber and fuel flow in said combustion chamber, said method
further including the steps of: providing a processor in signal
communication with said pressure transducers and said actuators,
wherein said processor receives said measured pressure fluctuation
signals and performs said associating step; and upon associating a
measured pressure fluctuation with an aerodynamic instability,
signaling said actuators from said processor to adjust engine
operation to minimize the occurrence of said aerodynamic
instability.
10. The method of claim 9 wherein said processor providing step
includes locating said processor remotely from said engine.
11. The method of claim 9 wherein said aerodynamic instability is
rotating stall.
12. The method of claim 9 wherein said aerodynamic instability is
surge.
13.-20. (canceled)
Description
BACKGROUND OF THE INVENTION
[0001] 1. Field of the Invention
[0002] The present invention relates to control of a compressor in
a gas turbine engine and, more particularly, to control of a
compressor by detecting and compensating for aerodynamic
instabilities.
[0003] 2. Description of the Related Art
[0004] With the introduction of the gas turbine engine, the speed
and reliability of air travel has improved significantly. The gas
turbine engine also known as a turbo-jet engine provides propulsion
through the acceleration of a stream of air or gas which is
expelled at a high velocity. The typical turbo-jet engine includes
three basic functional elements a compressor for gathering and
pressurizing the air, a combustor chamber for heating the already
pressurized air and a turbine for translating the energy released
from the pressurized and heated air into mechanical energy and
thrust to propel the aircraft forward. While jet engine technology
has advanced one of the safest and fastest growing markets for mass
transportation, the technology still suffers from problems caused
by rotational stall and surge caused by changes in the air flow
rates through the compressor. Such problems can be magnified by
environments where the speed of the engine and the air speed in
which the engine operates are changed. While providing an optimum
operating environment can reduce the occurrence of stall and surge,
these same problems have arisen in gas turbine engines implemented
in the power generation field where the engine are operated at
generally constant speeds with a controlled air flow
environment.
[0005] The problem is that stall and surge are more likely to occur
when the engine is operated at or near its optimum operating speed.
One solution to the stall and surge problem has been to implement a
feedback and control system that uses measured pressure or pressure
and temperature characteristics to detect when conditions relating
to stall and surge are about to occur. The measured signals are
processed by a control circuit that detects a stall or surge
condition and adjust the engine operating parameters to eliminate
the measured conditions indicative of a stall or surge in the
engine. While such solutions have worked well in implementations
relating to turbine engines relating to power systems, such
solutions have been hampered in the use of such solutions for jet
engines. One problem has been the installation of sensors to detect
the air flow conditions. The operational environment of the turbine
engine causes the sensors to be subjected to extreme temperatures
and vibrational conditions. While the sensors in gas turbines for
power generation and the like may be mounted in a way to isolate
the sensor from such harsh conditions, the turbine engines used in
jet aircraft have weight and aerodynamic considerations that make
such techniques impractical. Compounding the problems in turbine
engines for jet aircraft has been the advances made in the
introduction of aluminum and composite materials into the jet
engine design. Such materials help to incrementally increase
efficiency and reduce weight; however, such materials have also
increased vibration encountered in the engine. The result of these
advances is the operating conditions in which the sensors must
operate have become more severe.
[0006] Thus a need exists for a way to implement a surge detection
system in a jet aircraft which improves the operational parameters
of the engine without sacrificing the aero dynamic and weight
considerations in the design.
SUMMARY OF THE INVENTION
[0007] The present invention relates to a system for detecting
aerodynamic instabilities in a jet turbine engine having a pressure
transducer mounted in the engine by welding. The pressure
transducer, welded to a circuit in signal communication with a
controller, is adapted to send measured pressure readings from air
in a combustion chamber to the controller. The controller, located
in spaced apart relation from the engine, is adapted by software to
detect pressure patterns from the pressure signals generated by the
transducer that are indicative of a stall or surge. A series of
fuel and air valves located with compression and combustion
chambers of the engine are in signal communication with the
controller. The controller in response to detecting pressure
signals indicating a stall or surge is operative to signals in the
valves to change the air flow or speed to reduce the possibility of
a stall or surge.
BRIEF DESCRIPTION OF THE DRAWINGS
[0008] Other objects and advantages of the invention will become
apparent from the foregoing detailed description taken in
connection with the accompanying drawings, in which
[0009] FIG. 1 is a partial, cut-away view of a turbofan jet
engine;
[0010] FIG. 2 is a diagram illustrating the effects of rotating
stall;
[0011] FIG. 3 is a diagram of air flow in a jet engine during a
stall;
[0012] FIG. 4 is a diagram of air flow in a jet engine during a
surge;
[0013] FIG. 5 is a diagram illustrating the effects of surge;
[0014] FIGS. 6 A&B are graphs correlating speed to surge for
uncompensated and compensated operation;
[0015] FIG. 7 is a block diagram of a jet engine fitted with a feed
back and control logic diagram;
[0016] FIG. 8 is a block diagram illustrating the location of
sensors in a compression chamber;
[0017] FIG. 9 is a block diagram of a direct mounted sensor;
[0018] FIG. 10 is a block diagram of a semi-infinite sensor
mounting;
[0019] FIG. 11 is a block diagram of a pitot probe
configuration;
[0020] FIG. 12 is a block diagram of an alternate pitot probe
configuration.
[0021] FIG. 13 is a diagram of silicon based pressure
transducer;
[0022] FIG. 14 is an isometric top view of two wafers aligned for
bonding;
[0023] FIG. 15 is a diagram of a pressure transducer;
[0024] FIGS. 16A & B are diagrams of an assembled pressure
capsule;
[0025] FIG. 17 is a diagram of an ultrahigh temperature leadless
pressure transducer;
[0026] FIG. 18 is a graph of zero output during temperature
cycling;
[0027] FIG. 19 is a graph of full scale output over repeated
cycles;
[0028] FIG. 20 is a graph of pressure v. output voltage at various
temperatures;
[0029] FIG. 21 is a graph of sensor performance up to 900.degree.
F.;
[0030] FIGS. 22-24 are graphs of zero to full scale output;
[0031] FIGS. 25-27 are graphs of output vs. pressure
performance;
[0032] FIGS. 28-30 are graphs of the changes in zero output and
full scale output;
[0033] FIG. 31 is a diagram of a frequency response test
set-up;
[0034] FIGS. 32A & B are graphs of transducer response at
650.degree. F. subject to sinewave excitation;
[0035] FIG. 33 is a diagram of a test engine;
[0036] FIG. 34 is a diagram of a pressure transducer operational
test configuration;
[0037] FIG. 35 is a graph of a spectrum analyzer output at
idle;
[0038] FIG. 36 is a graph of a spectrum analyzer output at high
power;
[0039] FIGS. 37-40 are comparative graphs of test data acquired at
various power settings and temperatures;
[0040] FIG. 41 is a graph of power spectral density by frequency
computed from the data illustrated in FIGS. 37-40;
[0041] FIGS. 42A and B are graphs of signals during load
transients;
[0042] FIGS. 43A and B are graphs of transducer calibrations;
[0043] FIGS. 44A and B are graphs of pressure and temperature
during a first test;
[0044] FIGS. 45A and B are graphs of pressure and temperature
during a second test;
[0045] FIG. 46 is a graph of test data acquired at various power
settings during the first test of FIG. 44;
[0046] FIG. 47 is a graph of test data acquired at various power
settings during the first test of FIG. 45;
[0047] FIG. 48 is a graph of power spectral density by frequency
computed from the data illustrated in FIG. 44; and
[0048] FIG. 49 is a graph of power spectral density by frequency
computed from the data illustrated in FIG. 45.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT(S)
[0049] With reference to the figures for purposes of illustration,
the present invention is embodied in a jet engine 20 (FIG. 1). A
typical jet engine design suitable for use in commercial aviation
is two or three spool, turbo fan which generally includes a low
pressure compressor 22 that collects air to improve thrust and
feeds a portion of the collected air into an intermediate pressure
compressor 24 or booster stages, in the case of a two spool design.
Pressurized air from the intermediate (or Booster) pressure chamber
24 is subsequently fed to a high pressure chamber 26 which then
feeds the highly pressurized air into the combustion chamber 28. As
used throughout this description the term "intermediate chamber"
may also refer to "booster chamber." The pressurized air is then
mixed with fuel from fuel injectors 30 in the combustion chamber 28
and ignited. The pressurized and heated air is then fed into a
turbine region defined by a series of turbine blades 32-34 that are
rotated by the exiting air and as the air exits it results in
thrust. With the turbines 32-34 connected on the shaft to
respective compressor stages 36-38 in the front of the engine, the
turbines rotated by the exiting air result in the rotation of the
compressors. This basic jet engine design has been the principle
form of propulsion used in most commercial aviation today. More
recently through computer modeling and experimentation, it has been
realized that dynamic instabilities in the air flow through the jet
engine can lead to rotational stall and surge. Advantageously, the
present invention incorporates high temperature, vibration
compensated pressure transducers in the jet engine to measure air
pressure in the booster/intermediate and/or the high compression
chambers which then measures and communicates signals indicative of
the measured pressure to a feedback and control circuit. It should
be understood by those skilled in the art that there are a number
of turbo fun engine designs with a variety of compression chamber
designs. While a three stage chamber is used as the engine for this
description, the vibration compensation solution disclosed herein
is fit for use in a jet engine having any type of compression
chamber configuration.
[0050] The occurrence of rotational stall can be brought on by a
number of factors in including changes in the operational speed of
the engine as well as wind and temperature changes in the
surrounding air in which the engine is operating. With reference to
FIGS. 2 and 3, rotating stall is when a portion 40 of the air 42
circulating about the circumference of the engine wall begins to
experience a temporary back flow or reverse flow 40 in the
direction of air passing through the compressor. While only a
portion of the overall air is pushed back, this reverse flow begins
to disrupt the flow of air in the compression chambers. In rotating
stall, the reverse flow actual rotates about the circumference of
the compression chamber and begins to oscillate at a rate of
typically 50 to 100 Hz on large engines. As the effect becomes more
pronounced rotating stall can lead to a surge (FIGS. 4 and 5),
which is where the air flow 42 actually reverses completely
resulting in a wave in which air back flows toward the front 43 of
the engine. When stall occurs, the solution to overcome a stall
condition is usually throttle back or to shut down and restart the
engine.
[0051] With reference to FIGS. 6A-B, stall and surge are related to
the relationship of pressure rise and mass flow during varying
operating speeds 44. Even with the highest degree of tolerances in
the engine design the peak operating efficiency of an engine
without control can be 20-25% below optimal efficiency. This is due
to the occurrence of rotational stall and surge which may appear as
the engine approaches its optimal operating conditions. In studies
conducted on the ground, it has been shown that the occurrence of
surge can actually be shifted through feedback and control systems
to allow for an actively stabilized operating point at the optimal
operating pressure rise and mass flow.
[0052] A sample feedback and control system 60 (FIG. 7) may include
pressure sensors 62 located in or about the intermediate/booster or
high pressure compressor which provides feedback to a control law
device 64 such as a microprocessor located remotely from the high
pressure and temperature operating conditions of the engine. The
control law device 64 operates using known control law parameters
to interpret the measured pressure signals and in turn operates the
actuators 66 which may be valves to control either the air flow in
the engine or fuel flow or motors to adjust the angle of stator
vanes in anticipation of a stall or surge condition to adjust the
operating conditions to eliminate such a possibility. Control law
devices of the type suitable for this purpose may operate as
separate processor devices or may be included in existing control
logic devices adapted to monitor the engine performance. The
software for performing the control law uses a conventional Fourier
decomposition logic to detect air flow patterns measured from
pressure transducers located in the intermediate/booster and/or
high pressure compressors of the gas turbine engine. The software
is not a simple unweighted FFT, but rather curve fitting logic that
account for variations in under sampling and over sampling of the
measured characteristics. Control Law formula of the type suitable
for this purpose includes the control law logic outlined in
"Compression System Stability and Active Control" (2001) by J. D.
Paduano, E. M. Greitzer and A. H. Epstein which is incorporated
herein by reference.
[0053] In acquiring the information needed for the control law
device to accurately predict the occurrence of stall or surge, a
plurality of pressure transducers 70-77 (FIG. 8) are placed about
the circumference 74 of either the intermediate/booster or high
pressure compression chamber. The number of pressure transducers
70-77 and placement of the pressure transducers needs to be
sufficient to measure the waveform variations 78-80 in air flow
brought on by the occurrence of rotational stall. When the
transducers are placed in equal spaced apart relation it will be
appreciated that the number of transducers needs to be sufficient
to measure at least two points in the cycle of a rotational stall
waveform 78, 79 or 80. It will be appreciated that the number of
pressure transducers may be reduced by spacing the transducers at
varying spaced apart points (not shown) about the circumference to
ensure that varying frequencies are measured.
[0054] The pressure transducers may be installed in various
conventional installation configurations including, but not limited
to:
[0055] 1. Direct Mount 100 (FIG. 9) or Semi-Infinite Line Mount 110
(FIG. 10) generally referred to as a wall static installation with
the pressure transducer 102 either directly mounted in the wall 104
of the compressor casing with a flush sensing diaphragm 106 or
remote from the static tappings 112 using a non-resonate pipe
system 114.
[0056] 2. Embedded Transducer Mount 120 (FIG. 11) or Semi-Infinite
Line Mount 130 (FIG. 12) generally referred to as a pitot probe 122
mounting in which the pressure transducer 124 is either mounted
within a pitot probe to measure air pressure or the pressure
transducer 132 is mounted remotely using a non-resonant pipe system
134 to draw air from the pitot probe 136.
[0057] It will be appreciated by those skilled in the art that the
wall static installation is presently preferred as the pitot probe
would introduce objects into the air flow of the engine and the
consequences of such objects on the air flow have not yet been
fully studied.
[0058] The use of such systems has been known to work in
laboratories and in the fields of power generation where the
attachment of pressure transducers can be isolated from the harsh
environment of the combustion turbines. However, such
implementations cannot be easily implemented in combustion turbines
used for jet engines as the air dynamic properties of the gas
turbine engine must be maintained internally and externally to the
intermediate/booster and high pressure combustion chambers as the
flow external from these chambers (by-pass airflow) is used to
provide additional thrust to the aircraft.
[0059] The selection of the sensor type and location is a critical
factor in determining the effectiveness and practicality of an
engine surge and stall control system, as is the selection of the
actuator type and the algorithms used to process the data from the
sensors. A review of the most recently published material in the
field of active control of surge and stall in axial flow
compressors concludes that the most widely used physical parameter
to monitor the stability of a compressor is pressure, although the
measurement of gas flow using hot wire anemometers and the
measurement of gas temperature using high response thermocouple
probes have been used successfully.
[0060] The high response pressure data which is generated by the
dynamic pressure transducer is processed using one of many
proprietary algorithms in order to predict or detect the onset of
stall and surge. Although the operation and logic of the algorithms
described in the technical publications vary considerably, the data
requirements from the pressure sensors appear to be remarkably
similar.
[0061] The pertinent characteristics which are desired of a
compressor mounted stall and surge pressure sensor are high
sensitivity (ability to detect 70 Pa (0.01 psi) peak to peak
fluctuations), stability of sensitivity with temperature and time
(+5% to 10% FS) and the ability to survive in an extremely hostile
environment (operating ambient temperatures and transients between
-54.degree. C. and 400.degree. C. (-65.degree. F. and 750.degree.
F.) and vibration levels of typically 30 g rms between 50 Hz and 18
kHz). The pressure transducer installation should also have
sufficient bandwidth to measure frequencies between 100 Hz and 1
kHz for large gas turbines and between 500 Hz and 8 kHz for small
gas turbines with negligible phase shift. During surge conditions,
the pressure transducer must survive gas path pressure and
temperature transients of up to 3.4 MPa (500 psi) and 1000.degree.
C. (1830.degree. F.) for several seconds. Finally, if active surge
and stall control systems are to be applied to production civil and
military gas turbine engines in the future, the reliability and
cost of the dynamic pressure transducers must be competitive with
the pressure transducers currently used to measure oil, fuel, air
and hydraulic pressures on airframes and engines.
[0062] In order to understand the development which has led to the
creation of the leadless pressure transducer, it is relevant to
consider the original silicon-on-insulator pressure capsule
design.
[0063] The heart of the piezoresistive SOI pressure sensor 140
(FIG. 13) is a silicon diaphragm 142 which is supported upon a
Pyrex glass pedestal 144 in such a manner as to enable a pressure
differential to be applied across the diaphragm without introducing
a mounting strain in the diaphragm. An "anodic" molecular bond 146
is used to attach the silicon diaphragm to the glass pedestal which
ensures a very stable, permanent assembly without the use of glues
or adhesives. Piezoresistive silicon strain gauges 148 are
integrated within the silicon diaphragm structure, but are
electrically isolated by a SO.sub.2 barrier 150 from the silicon
diaphragm 142. The piezoresistors 148 measure the stress in the
silicon diaphragm 142 which is a direct function of the pressure of
the media. The silicon diaphragm 142 is usually thinned in selected
areas underneath the piezoresistors by anisotropic chemical etching
in order to increase the pressure sensitivity of the diaphragm.
[0064] In a conventional arrangement, the piezoresistors are
connected electrically via metallic interconnections to form a
fully active Wheatstone bridge. At the corners of the diaphragm are
placed five gold bond wires (not shown) which are ultrasonically
ball bonded to the diaphragm metallization and are used to
connected electronically to the bridge. Under extreme conditions to
temperature and vibration, the ultrasonic agitation used to form
the ball bonds causes abrasion to take place during the welding
process and allows microscopic holes to develop in the platinum
metallization through which, at high temperatures, the gold can
migrate and form a gold-silicon eutectic which causes the leads to
fail. In addition, the pressure media is in direct contact with the
stress-sensing network, leadouts and interconnects which at high
temperatures and in the presence of aggressive chemicals can fail.
The key elements in the design of a ruggedized pressure sensor is
the elimination of the gold bond wires and the protection of the
sensing elements from corrosive environments at thigh temperatures,
hence the reference to the new sensor capsule as the "leadless"
design.
[0065] The leadless sensor capsule 152 (FIG. 14) is comprised of
two main components, the sensor chip 154 and the cover wafer 156
which are eventually assembled to form the pressure capsule.
[0066] The sensor chip is manufactured from two separate wafers.
First a carrier wafer is fabricated which forms the mechanical
structure, the diaphragm. The second wafer is referred to as the
sacrificial wafer on which is defined the areas which the high
conductivity P.sup.+ piezoresistive strain gauges occupy. After
oxidizing the carrier wafer to form an electrically insulating
layer over its surface, the two wafers are bonded together using a
Diffusion Enhanced Fusion bonding (DEF) process. The bond is a
direct chemical molecular bond between the piezoresistive P.sup.+
regions and the silicon oxide and uses no adhesive or additional
components. Once the bond is formed, the non-doped areas of the
carrier wafer are selectively removed chemically. The
piezoresistive P.sup.+ regions are now permanently bonded to the
dielectrically isolated carrier wafer in which the diaphragm is now
micromachined. In order to optimise the mechanical performance of
the force collector, the diaphragm is formed in the shape of a
picture frame.
[0067] FIG. 15 shows a view of the sensor chip 160 with the four
piezoresistive gauges 162 strategically positioned inside the
"picture frame" shaped diaphragm 164 and connected in a Wheatstone
bridge circuit. The entire sensing network is P.sup.+ and there are
separations between the contact regions of the bridge, Metal is
deposited to form ohmic contacts to the P.sup.+ regions located
inside the large contact regions. There is also a rim of P.sup.+
material 166 around the periphery of the sensor chip. When the
cover wafer is assembled to the sensor chip, and hermetic seal is
formed between the cover and this area of P.sup.+ material thus
protecting the stress sensing network and all the electrical
interconnections from the harsh environmental conditions.
[0068] The cover wafer maybe manufactured from either silicon or a
Pyrex glass to the same dimensions as the silicon wafer. Four holes
are drilled in the cover, one in each corner, which align with the
metallised contact pad areas. A recess is also created in the
centre of the cover wafer to allow the diaphragm to deflect freely
when assembled.
[0069] The sensor chip and the cover wafer are then assembled using
an electrostatic bond. FIG. 14 shows a top isometric view of the
components just prior to sealing. Once the two wafers have been
bonded, only the metallised leadout pads are exposed whilst all the
gauges and electrical interconnections on the sensing side of the
silicon chip are sealed by the cover. Thus the active portion of
the pressure sensor is hermetically isolated.
[0070] To avoid the use of gold ball bonds and fine gold wires, a
high temperature metal frit is used to provide the electrical
connection between the sensing chip and a specially designed
header. The frit is a mixture of high conductivity metal powders in
appropriate physical form and glass and is used to fill the holes
in the cover wafer after it is bonded to the sensor chip.
[0071] The specially designed header 170 contains a group of four
hermetically sealed pins 172 protruding from its surface which are
spaced so as to fit the holes drilled in the cover wafer. FIGS. 16A
and B show a section of the assembled pressure capsule 176 and also
a section of the pressure capsule 176 mounted in the header 170.
The pressure capsule 176 is bonded to the header at a high
temperature using a non conductive glass frit 178, during which
process the metal frit 178 in the cover wafer holes melts and
creates low resistance electrical connections between the header
pins and the metal contact pads on the sensor chip 179.
[0072] After this firing process, only the non-active side of the
diaphragm is exposed to the pressure medium. The small ball bonded
gold leads have been eliminated and the entire sensor network and
contact areas are hermetically sealed from the environment and the
pressure media.
[0073] The hermetically sealed pressure sensing capsule bonded to
the header is the starting point for the assembly into a pressure
transducer. Typically most transducers must be attached to a
mounting surface which is exposed to the pressure media, frequently
by means of a threaded port. In addition, the header pins must be
electrically connected to a high temperature cable assembly without
the use of solder joints which may fail at high temperatures. The
high temperature cable assembly must also contain material which
will provide electrical insulation between individual leads, whilst
the interconnects between the header and the cable as well as the
cable itself must be strong enough to withstand the mechanical
stresses of handling. The package is completed using a building
block approach and FIG. 18 shows the assembly of a ultra high
temperature leadless pressure transducer 180.
[0074] A sleeve 182 is welded between the first header and a second
header. A minerally insulated (MI) cable containing nickel wires is
used to interconnect to the pins from the first header and the
exposed leads from the first header are welded to the second header
to ensure low resistance electrical connections between the leads
of the MI cable and the header leads.
[0075] The header/MI cable assembly is then inserted into a port
184 and welded to the port. At the end of the port is a tubulation
186 which is crimped to retain the MI cable.
[0076] A cover sleeve (not shown) is then assembled over the MI
cable to give additional support and is welded to the rear of the
cover which in turn is welded to the port 184.
[0077] This design of assembly results in the transducer being
totally hermetically sealed from any atmospheric contamination or
oxidation. Every single internal metallised surface such as metal
to silicon and metal to glass frit, header pins to header tubes,
header pins to MI cable wires and even the mineral insulation
itself is hermetically sealed from the atmosphere. In addition the
welding of the sleeve to the port together with the addition of the
third header greatly increases the structural integrity of the
entire electrical interconnect system and reduces the chances of
any damage in severe environments.
[0078] The first generation of leadless transducers manufactured
(five devices) have been tested in the laboratory with the
following results. FIG. 18 shows the change in zero output during
repeated temperature cycling between room temperature and
455.degree. C. (850.degree. F.). This demonstrates that exposure to
high temperatures has negligible effect on the internal electrical
connections and contacts. A few ohms change in a contact resistance
would result in changes in the output of many millivolts. All
observed changes in output were less than 2 mV. FIG. 19 plots the
full scale output at 455.degree. C. (850.degree. F.) for two
sensors over repeated cycles. Stable and repeatable outputs were
observed throughout this study.
[0079] FIG. 20 shows the pressure v output voltage performance
measured at room temperature, 177.degree. C. (350.degree. F.),
343.degree. C. (650.degree. F.) and 455.degree. C. (850.degree. F.)
for one of the sensors. There is a small element of zero shift but
the unit is very linear and exhibits a repeatable span shift of
approximately 2-3%/100.degree. C. (1-2%/100.degree. F.). FIG. 21
shows sensor performance up to 482.degree. C. (900.degree. F.) for
another one of the tested sensors. Linearity and span shift remain
virtually identical.
[0080] In summary, the devices appear to have less than 0.02% F.S.
non-linearity and no measurable hysteresis up to temperatures of
343.degree. C. (650.degree. F.). At temperatures of 454.degree. C.
(850.degree. F.) the non-linearity increases to around 0.1% F.S.
but a static error band of better than 0.15% F.S. can be expected.
All units tested exhibited only minor changes in performance
characteristics after repeated exposure to high temperatures. When
the units were compensated, span and zero shifts of less than 1%
F.S. over the temperature range from room temperature to
400.degree. C. (750.degree. F.) were achieved.
[0081] The latest generation of leadless transducers manufactured
has been tested in the laboratory with the following results. FIGS.
22, 23, and 24 show the full scale output and zero output
performance measured at room temperature, 200.degree. F.,
400.degree. F., 600.degree. C., 800.degree. F. to 900.degree. F.
(480.degree. C.) for three of the non-compensated sensors. There is
a small element of zero shift with temperature but the units all
exhibit a repeatable span shift of approximately (2.5%/100.degree.
F.) over the entire temperature range up to 900.degree. F. Zero
output shift with temperature and span shift remain virtually
identical for all three tested sensors.
[0082] FIGS. 25, 26, and 27 show output vs. pressure performance
for two compensated transducers. To summarize the test results, the
devices appear to have less than 0.02% F.S. non-linearity and no
measurable hysteresis up to 900.degree. F. Units tested exhibited
only minor changes in output vs. pressure performance
characteristics over the temperature range. FIGS. 28, 29 and 30
show the change in zero output and full scale output for three
compensated transducers during repeated temperature exposure to
900.degree. F. (480.degree. C.). This demonstrates that exposure to
high temperatures has negligible effect on the internal electrical
connections and contacts. A few ohms change in a contact resistance
would result in changes in the output of many millivots. All
observed changes in output were less than a few mV. These
compensated units were exposed to 900.degree. F. for over 3 hours
without any noticeable degradation in performance. In summary, the
span ad zero shifts of less than 1% F.S. over the temperature range
from room temperature to 900.degree. F. (480.degree. C.) were
achieved.
[0083] The design of the high temperature sensor is such that it
should have high frequency response characteristics similar to
those of more familiar, low temperature capability Kulite sensors.
To very this experimentally, a pulsed air apparatus was set up in
an oven.
[0084] The frequency response test set up 320 is shown in FIG. 31.
Large scale pressure primary pulsation at frequencies up to 400-500
Hz were generated by a water cooled, motor driven rotary valve 322
with an 1/4'' port. The valve was mounted immediately exterior to
an oven 324 containing the test transducer 326. About 15 cm of
1/4'' stainless steel line 327 connects the valve to the transducer
326, which is mounted on one leg of a T off the line. A second,
standard, lower temperature capability transducer 328 (Kulite model
XTC-190) is mounted in the opposite leg of the T. After passing by
the transducers, the flow exits the over through 15 cm of the line
330 to a manual throttle valve 332.
[0085] The response of both transducers was first established at
room temperature. The high temperature unit and low temperature
reference unit had essentially identical waveform shape and
frequency response. This verifies that the transducer response is
as expected. The reference unit was then removed and the test
repeated at elevated temperatures, after appropriate soak time.
[0086] An example of the transducer response at 650.degree. F.
subject to a nominally 250 Hz sinewave excitation is shown in
FIGS., 32A & B. The amplifier gain is 200. At the higher
frequency of 400 Hz, the wave form is less sinusoidal due to
resonance in the flow system. The second harmonic response is
clearly visible at 800 Hz. These tests are greatly constrained by
the limitations of the excitation mechanism and so do not fairly
illustrate the frequency response capabilities of the sensor, which
is many tens of kilohertz. The data do however, demonstrate nearly
ideal ac response through the range of interest for many gas
turbine active control applications.
[0087] The latest generation of dielectric isolated sensors have
been fabricated and evaluated which employ the Kulite leadless
design. The key features of the leadless design are the elimination
(of the gold bonding and gold lead wires) and the hermetic sealing
of the pressure capsule and the transducer assembly which will
enable these transducers to operate in the most hostile
environments.
[0088] Through experimentation a sensor of the type suitable for
use in the combustion engine of a jet aircraft has been found to be
a model nos. XTEH-7L and XTEH-10A pressure transducers manufactured
by Kulite Semiconductor Corp. of Leonia, N.J. The features of these
type of pressure transducers include fabrication for high
temperature and high pressure operating environments vibration
compensating features to distinguish false pressure variation
measurements caused by vibration from actual air flow vibrations as
well as a leadless installation in which the transducer can be
directly welded to leads that connect to the control law device to
thereby prevent open circuit conditions caused by vibrations acting
upon the solder contacts. The features of these transducers are
fully disclosed in U.S. Pat. Nos. 5,286,671, 6,293,154, 6,272,929,
5,955,771, 6,327,911 and 6,363,792 all assigned to Kulite
Semiconductor Corporation and all of which are incorporated herein
by reference. To establish the feasibility of these high pressure
and high temperature pressure transducers for use in a jet aircraft
combustion engine experiments were conducted on the Kulite XTEH-7L
to judge whether it can be used in the actually operating
environment of a combustion aircraft engine. The results of the
experiments are detailed below as follows:
[0089] Experimental Assessment for Gas Turbine Testing
[0090] Introduction
[0091] An uncooled Kulite XTEH-7L high temperature pressure
transducer was tested mounted on the combustor of a Rolls-Royce
(Allison) S250-C30 turboshaft gas turbine engine to demonstrate
transducer dynamic behavior in a realistic engine environment. A
low temperature transducer (XCQ-062) in a water-cooled casing was
mounted in parallel as a reference. At all temperatures tested (up
to 700.degree. F.), the high temperature transducer dynamic
response was very similar to that of the low temperature unit.
[0092] Installation
[0093] A Rolls-Royce 5250-C30 engine (FIG. 33) was set up for
experiments exploring the active control of surge. To raise the
compressor operating line and facilitate surge studies, the engine
can be equipped with flow blockages between the compressor
discharge and combustor entrance (which requires minor engine
disassembly for modification). In addition, de-ionized water can be
injected into the compressor discharge to further raise the
operating line in a controlled manner during an experiment while
keeping the turbine entry temperature within allowable limits.
Dynamic fluid forcing of the engine and feedback control is
implemented with compressed air inbleed through the boundary layer
suction slot around the compressor inducer. The bleed is at the
compressor diffuser discharge pressure but is supplied by
laboratory compressors in these experiments. A high frequency
control valve powered by a Moog actuator modulates this flow at
large amplitude up to a frequency of 400 Hz.
[0094] Dynamic and static measurements within the engine are
accomplished with arrays of Kulite low temperature XCQ transducers
340 mounted upstream, along the compressor flow path, and in the
combustor 342 as shown in FIG. 33. Since the frequencies of
interest for surge are less than 1 kHz, the transducers are mounted
on 0.020'' (0.5 mm) diameter SS tubing stubs less than 10'' (12 cm)
long. The transducers are mounted in low internal volume,
water-cooled housings for thermal stability and long-term
protection in the hot engine environment.
[0095] With reference to FIG. 34, the Kulite XTEH-7L-190-100A high
temperature pressure transducer 350 is mounted on a 1/4'' tubing T
fitting 352 attached to a combustor drain plug 354. On one side of
the T fitting 352 is a reference Kulite XCQ-062-250G pressure
transducer 356 in a water-cooled housing 358. The XTEH-7L pressure
transducer 350 is mounted to the other leg of the fitting. Because
the ambient temperature is low in the laboratory environment,
unlike a typical aircraft installation, an electric heater (not
shown) has been attached to the tubing fitting and the XTEH-7L
insulated to achieve temperatures more typical of installation in
an advanced large engine. A thermocouple monitors the transducer
temperature.
[0096] The data system for the transducer consists of Pacific
Scientific instrumentation amplifiers feeding a 16-bit A/D system.
Excitation voltage was 15 V for both transducers. The analog signal
is unfiltered, and the sampling rate used was between 5 and 20 kHz
(see individual plots).
[0097] Data and Discussion
[0098] Tests were performed at various temperatures between
250.degree. F. and 700.degree. F., at both idle and high power
operating conditions (see FIGS. 35-42). Mean pressures ranged from
20 to 90 psi, while perturbations were between 2 and 10 psi peak to
peak. Broadband combustion noise was recorded, with several
spectral peaks between 0 and 5000 Hz. Although there were slight
differences in the spectral content of the two transducers, their
signals were overall quite similar. In fact, the gain from the XCQ
transducer was used to reduce the XTEH data, as a surrogate for an
in site calibration. Offsets were chosen so that the mean values of
the traces are identical in the plots shown here.
[0099] Two data acquisition methods were applied. The first was
simply to use a spectrum analyzer with storage capability. FIGS. 35
and 36 show the output of the spectrum analyzer for 2 different
conditions--idle power (20 psi mean) at 700.degree. F., and high
power (56 psi mean) at 500.degree. F. The two primary spectral
peaks below 1000 Hz are captured by both transducers, as are
secondary spectral peaks between 1 kHz and 5 kHz. In the region
between 1500-2000 Hz there are differences between these plots.
Although the signal power in this frequency range is 2 to 3 orders
of magnitude lower than the peak values, it is believed that this
discrepancy is due to acoustic effects in the longer lines leading
to the XCQ transducer (signal power is larger for the XCQ
transducer in this range).
[0100] The second data acquisition method was an 8-channel DSP-base
data acquisition system sampling between 5 and 20 kHz. FIGS. 37
through 40 compare raw traces from this data acquisition system for
various power settings and temperatures. These traces also show
very similar features, with the traces overlaying one another in
each case. Spectra can also be computed from this data; an example
is shown in FIG. 41. Finally, in FIGS. 42A and B, a power transient
is recorded and a side-by-side comparison of the two transducers is
shown. Note that the XCQ calibration is not exactly right over the
full range of the transducer, but that both transducers basically
measure the entire transient in a similar way.
[0101] Kulite XTEH Transducer Tests on a Gas Turbine Jet Engine
[0102] Introduction
[0103] Two uncooled Kulite XTEH type high temperature pressure
transducer (XTEH-7L-190-200A and XTEH-10AC-190-200A) were tested
mounted on the combustor of a Rolls-Royce (Allison) S250-C30
turboshaft gas turbine engine to demonstrate transducer dynamic
behavior in a realistic engine environment (FIG. 33). A low
temperature transducer (XCQ-062-250G) in a water-cooled casing was
mounted in parallel as a reference. The XTEH 7L transducer was
tested up to 700 deg F., and the XTEH-10AC transducer was tested up
to 900 deg F.
[0104] Installation
[0105] A highly instrumented Rolls-Royce 5250-C30 engine was used
for the tests described here. Static and dynamic measurements
within the engine are accomplished with arrays of Kulite low
temperature XCQ transducers mounted upstream, along the compressor
flow path, and in the combustor. Since typical frequencies of
interest in engine dynamics are less than 1 kHz, the transducer are
mounted on 0.020'' (0.5 mm) diameter stainless steel tubing stubs
that are less than 10'' (12 cm) long. The transducers are mounted
in low internal volume, water-cooled housings for thermal stability
and long-term protection in the hot engine environment.
[0106] The Kulite high temperature pressure transducers were
mounted on a 1/4'' tubing T fitting attached to a combustor drain
plug. On one side of the T fitting is a reference XCQ-062-250G
pressure transducer in a water-cooled housing similar to the
configuration of FIG. 34. The XTEH pressure transducers were
mounted to the other leg of the fitting. Separate tests were
conducted with the XTEH-7L and the XTEH-10AC mounted in the same
fitting. Because, unlike a typical aircraft installation, the
ambient temperature is low in the laboratory environment, an
electric heater has been attached to the tubing and the XTEH-7L
insulated to achieve temperatures more typical of installation in
an advanced large engine. A thermocouple monitors the transducer
temperature.
[0107] The data system for the transducer of Pacific Scientific
instrumentation amplifiers feeding a 16 bit A/D system. Excitation
voltage was 15 V for both transducers. The analog signal is
unfiltered, and the sampling rate used was 5 kHz.
[0108] Two engine runs were conducted (one for each transducer).
The only difference between the runs was that the maximum
temperature tested for the XTEH-7L transducer was 700 deg F., and
the maximum temperature for the 10AC transducer was set at 900 deg
F. Both runs consisted of first running engine at idle, taking
unsteady data with the Kulites heated to 500 F. This was then
followed by a spool-up to max throttle, which causes the spool
speed to accelerate to 70%, and recording the transient. The engine
was then tested at various speeds (between 70% and 90%) and various
transducer temperatures. Since heating of the transducers was not
controlled, and engine run time was limited, temperature of the
probe varied by as much as 5 deg F. during the sample intervals;
this was more pronounced during XTEH-10AC testing due to the large
temperature change that was induced by heating. During XTEH-7L
testing, the rate of temperature increase was never greater than
0.5 deg/sec.
[0109] Calibration
[0110] Since the water-cooled reference transducer was a
differential transducer, calibration using a vacuum pump on the
back side of the transducer was performed. This calibration was
subject to two sources of inaccuracy: the calibration range was 0
to 14 psia while the operating range was up to 125 psia, and the
vacuum line was leaky, so that the minimum pressure in the
calibration was not very accurate. Thus the XCQ calibration was
relatively poor. To allow comparison of the transducer signals,
however, this transducer was then taken as the reference
transducer, and data taken during the engine run (at various
temperatures and combustor pressures) was used to drive calibration
constants for the XTEH transducers. The resulting calibration
curves for the XTEH transducers are shown in FIGS. 43A and B.
[0111] Test Results
[0112] Using these pseudo-calibrations to make the transducer
outputs comparable, a summary of the runs can be made. By plotting
each 10-second transient in sequence, FIGS. 44A and B and 45A and B
show a sequential history of the pressure outputs and the
temperatures tested. Note that there are gaps of several minutes
between each data set, so that entire run is not really as short as
the sequence shown; concatenation of the time histories is
asynchronous with the actual measurements. Both transducers track
the cooled transducer outputs well up to 700 or 800 deg F.;
however, at the highest temperatures tested the 10AC transducer
failed.
[0113] It is apparent in the FIGS. 44 and 45 that both transducers
track the low frequency transients and oscillations in the same
way. To look at higher frequency oscillations, very short sequences
of data are plotted in FIGS. 46 and 47. Finally, spectral analysis
of the data was performed; the results for typical data sets are
shown in FIGS. 48 and 49. In general, the high temperature
transducers appear to have a lower noise floor, allowing them to
resolve frequency peaks more distinctly.
[0114] While the present invention has been described in connection
with what are presently considered to be the most practical and
preferred embodiments, it is to be understood that the invention is
not to be limited to the disclosed embodiments, but to the
contrary, is intended to cover various modifications and equivalent
arrangements included within the spirit of the invention, which are
set forth in the appended claims, and which scope is to be accorded
the broadest interpretation so as to encompass all such
modifications and equivalent structures.
* * * * *