USOO6637932B2

(12) United States Patent (10) Patent No.: US 6,637,932 B2

Badeer (45) Date of Patent: Oct. 28, 2003

(54) ENGINE THRUST BEARING CONDITION 4,502,043 A 2/1985 Moore ........................ 374/144
MONITORING METHOD 4,620,185. A * 10/1986 Plahmer ................... 73/119 R
4,878,184 A * 10/1989 Okada et al. ............... 374/169
(75) Inventor: Gilbert H. Badeer, Loveland, OH (US) 2.2- --2
A : 3. Heat - - - - - - - - - - - - - - - - - - - - - 3.:
COIan K . . . . . . . . . . . . . . . . . . . . .
4,982,347 A * 1/1991 Rackerby et al. ........... 374/137
(73) ASSignee: Seal Eley, mpany, 5,080,496 A 1/1992 Keim et al. ................. 374/144
chnectady, NY (US) 5,176,034 A * 1/1993 Hazony et al. ............... 73/597
5,319,963 A 6/1994 Benford ...................... 374/141
(*) Notice: Subject to any disclaimer, the term of this 5,803,608 A * 9/1998 Randoll et al. ............. 374/144
patent is extended or adjusted under 35 5,892,145 A * 4/1999 Moon et al. ..... ... 73/118.2
U.S.C. 154(b) by 0 days. 5.941927 A * 8/1999 Pfitz ........................... 374/144
6,050,087 A 4/2000 Kurihara et al. ............ 374/144
(21) Appl. No.: 10/214,294 6,095,946 A * 8/2000 Maguire et al. .............. 477/98
6,141,087 A * 10/2000 Viel ................... ... 356/35.5
(22) Filed: Aug. 7, 2002 6,360,616 B1 * 3/2002 Halliday et al. ......... 73/862.49
6,522,990 B1 * 2/2003 Rains et al. ................ 374/169
(65) Prior Publication Data
* cited by examiner
US 2002/0186747 A1 Dec. 12, 2002
Primary Examiner Diego Gutierrez
Related U.S. Application Data Assistant Examiner Yaritza Guadalupe
(74) Attorney, Agent, or Firm William Scott Andes;
(62) Division of application No. 09/399,196, filed on Sep. 17, Barbara Joan Haushalter
1999, now Pat. No. 6,439,767.
57 ABSTRACT
(51) Int. Cl." .................................................. G01K 1/08 (57)
(52) U.S. Cl. ....................... 374/141; 374/134; 374/144; A method is provided for monitoring engine thrust bearing
73/119 R; 73/117.4 temperature deviations. Initially, key engine parameters are
(58) Field of Search ................................. 374/141, 142, measured. A bearing temperature rise can be predicted based
374/144, 134, 102, 112, 169 on the measured parameters. The predicted rise value is
compared with the actual rise value to generate a compari
(56) References Cited Son value. The System is trained to respond to the compari
U.S. PATENT DOCUMENTS
Son Value to prevent catastrophic failure.
4,318,179 A 3/1982 Kure-Jensen et al. ......... 73/462 6 Claims, 2 Drawing Sheets

- 30
MEASURE KEY ENGINE 32
PARAMETERS

PRED CT SUMP TEMPERATURE

RSE BASED ON MEASURED 34
PARAMETERS

MEASURE ACTUAL SUMP

TEMPERATURE RISE 36

COMPARE PREDCTED
WITH ACTUAL 38

40

COMPARISON
< SET POINT

42
DEPLOY COMPARISON
ALARM SET
PONT
SHUT DOWN
ENGINE
U.S. Patent Oct. 28, 2003 Sheet 1 of 2 US 6,637,932 B2

s:
U.S. Patent Oct. 28, 2003 Sheet 2 of 2 US 6,637,932 B2

- 30
MEASURE KEY ENGINE
PARAMETERS 32

PRED CT SUMP TEMPERATURE

RSE BASED ON MEASURED 54
PARAMETERS

MEASURE ACTUAL SUMP

TEMPERATURE RISE 36

COMPARE PREDCTED
WITH ACTUAL 38

YES/COMPARISON
< SET POINT

44 42
DEPLOY NO 1COMPARISON YES
ALARM D SET
PONT
B? SHUT DOWN
46 ENGINE

FG. 2
 US 6,637,932 B2
1 2
ENGINE THRUST BEARING CONDITION DETAILED DESCRIPTION OF THE
MONITORING METHOD INVENTION
The present invention compares a predicted Sump tem
This is a divisional of Application No. 09/399,196, filed perature rise to an actual measured value from the engine.
Sep. 17, 1999, now U.S. Pat. No. 6,439,767. When the difference between predicted versus measured
BACKGROUND OF THE INVENTION reaches a predetermined setpoint, an alarm and/or engine
trip function is deployed. This technique can be built directly
The present invention relates to gas turbine engines and into the engine control Software to prevent catastrophic
particularly to engine thrust bearing condition associated engine failure.
with the gas turbine engines. Referring now to the drawings, there is illustrated in FIG.
Industrial engines on occasion can exhibit catastrophic 1 a partial cutaway view of a typical industrial engine 10.
Secondary engine damage upon incurring engine thrust The engine thrust bearing condition monitoring System and
bearing failure. Failures of thrust bearings can jeopardize the method herein can be applied to Such an engine 10. Several
Serviceability of engine high pressure turbine and compres 15 of the engine measured parameters are used to predict
Sor airfoils. This is particularly true when thrust bearing bearing temperature rise. This predicted bearing temperature
failure detection and Subsequent engine shutdown measures rise is compared to the measured value. When the difference
are not immediately executed. is greater than a predetermined value, automatic shutdown
General industry practice has been to monitor bearing prevents catastrophic engine failure.
lubrication Supply and Scavenge temperatures. Another form Continuing with FIG. 1, a high pressure turbine rotor
of protection is a chip detection System located at bearing (HPTR) 12 tends to load thrust bearing 14 in an axially aft
Scavenge lines. These Systems monitor the level of particles direction. Meanwhile, the high pressure compreSSor rotor
released by the bearing or associated Sump hardware. These (HPCR) tends to counter that axial load by generating a high
methods of protection primarily provide alarm and trip forward load. These two loads are not equal and are rela
commands to the control System. An alarm only mode 25 tively high as compared to other engine cavity preSSure
permits an operator to acknowledge and correct the detected contributors. These other engine cavities, such as P3 18, P4.8
fault by reducing engine load level. However, during 20 and HP recoup 22, are all part of this axial loading of the
extended engine operating periods, Subsequent to thrust thrust bearing. However, their individual axial load ampli
bearing failure, catastrophic Secondary damage can occur as tudes are far Smaller than HPTR and HPCR contributors.
Scavenge temperature rises very rapidly. The HP recoup cavity pressure 22 is also an adjustable
It would be desirable, then, to be able to monitor engine parameter. This adjustable feature is used to balance the net
thrust bearing temperature using a predicted Sump tempera bearing loads to within an acceptable window of operability.
ture change to prevent catastrophic Secondary engine dam The other element of the load equation is the engine 10
age. Speed. Increasing Speed levels tends to increase the thrust
35 bearing ball radial loads. This generates more heat which is
BRIEF SUMMARY OF THE INVENTION Subsequently reflected by the increased Sump Scavenge
The present invention proposes to utilize various mea temperature. Finally, the compressor rear frame (CRF) 24 is
Sured engine parameters to predict Sump temperature rise, a structural membrane that houses multiple components.
and then use the predicted Sump temperature rise to monitor Referring now to FIG. 2, there is illustrated a flowchart
engine thrust bearing temperature. The predicted Sump tem 40 diagram 30, illustrating the Steps for predicting and moni
perature rise is compared to the actual measured bearing toring engine thrust bearing conditions. Initially, at Step 32,
temperature, and when the difference is greater than a key engine parameters are measured. Key Select contributors
predetermined value, the engine control System can auto are high preSSure recoup cavity pressure, engine Speed,
matically shut the engine down. This would prevent a Station P4.8 pressure, Station P3 pressure, and Some constant
catastrophic engine failure. 45 parameter. When these four Select variables or parameters
A method is provided for monitoring engine thrust bear are used, the result is that a predicted temperature with a
ing temperature deviations. Initially, key engine parameters standard deviation of a mere 6.8 deg F. (3.8 deg C.) is
are measured. A bearing temperature rise can be predicted achieved. Of these four variables, three constitute the major
cavity preSSures that are technically the major drivers. The
based on the measured parameters. The predicted rise value 50 last variable is engine Speed, which can be expected to have
is compared with the actual, measured, rise value to generate a direct load contribution based on the physics of bearing
a comparison value. The System is trained to respond to the design.
comparison value to prevent catastrophic failure. At block 34, the Sump temperature rise is predicted, based
Accordingly, the present invention provides an effective on the measured parameters from block 32. The Sump oil
technique for monitoring engine thrust bearing temperature 55 temperature rise includes all the contributors that affect
using a predicted Sump temperature rise. The capability of Scavenge oil temp. More than one bearing may be in any
predicting Sump temperature rise and automatically tripping given Sump (roller and ball types of bearings). Likewise, the
the engine can prevent a catastrophic engine failure. environment Surrounding the Sump may also contribute to
Furthermore, this reduces engine rebuild costs and down the Sump Scavenge oil temp rise. Conversely, bearing tem
time. 60 perature rise is exclusively for that particular bearing in a
BRIEF DESCRIPTION OF THE DRAWINGS given Sump which experiences the measured temperature
rise. Subsequently, at Step 36, the actual Sump oil tempera
FIG. 1 is an engine illustrating the thrust bearing and other ture rise is measured. At block 38, the predicted value of
components relevant to predicting bearing temperature rise; block 34 is compared with the actual value of block 36. If
and 65 the comparison is less than a Set point A, StepS 32 through
FIG. 2 is a flowchart block diagram illustrating the Steps 38 are repeated, to decision block 40. If the comparison is
for predicting bearing temperature rise. greater than or equal to a Set point A, the program proceeds
 US 6,637,932 B2
3 4
to decision block 42. If the comparison is less than a Set modifications may be made to adapt a particular situation or
point B, an alarm is deployed at block 44, indicating the material to the teachings of the invention without departing
Sigma difference. If the comparison is equal to or greater from the essential Scope thereof. Therefore, it is intended
than a Set point B, the engine can be shut down at block 46 that the invention not be limited to the particular embodi
to prevent engine damage. This forces the operator/end user ment disclosed as the best mode contemplated for carrying
to physically inspect the engine for disclosure of failed out this invention, but that the invention will include all
components. embodiments falling within the Scope of the appended
A regression analysis can be used to predict the Sump claims.
temperature rise, as illustrated by the flowchart in FIG. 2. What is claimed is:
The Scavenge temperature delta is determined using con 1. A method for monitoring engine thrust bearing
Stants and measured engine parameters. The high pressure temperature, comprising the Steps of:
recoup cavity preSSure is multiplied by a recoup pressure measuring predetermined engine parameters,
coefficient to provide a recoup contribution value. The P4.8 predicting a bearing temperature rise based on the mea
cavity pressure is multiplied by a cavity coefficient to Sured parameters,
provide a P4.8 cavity contribution value. The engine speed 15
detected by rotational Velocity of the core engine is multi measuring actual bearing temperature rise,
plied by an engine Speed coefficient. This gives an engine comparing the predicted rise value with the actual rise
speed contribution value. The pressure at the PS3 cavity value to generate a comparison value, and
compressor-discharge pressure is multiplied by a coefficient. responding to the comparison value to help prevent
The Summation of these four engine parameters, each mul engine failure.
tiplied by its associated coefficient, is used to calculate the 2. A method as claimed in claim 1 wherein the Step of
predicted Sump temperature rise. Finally, a fixed constant responding to the comparison value comprises the Step of
Sump temperature rise can be added to the equation. The end automatically triggering a signal when the comparison value
result is a predicted Sump temperature rise. Alube oil Supply exceeds a predetermined Sigma level.
28 temperature can be Subtracted from the predicted Sump 25
3. A method as claimed in claim 2 wherein the predeter
temperature discharge/Scavenge to give a delta value. The mined engine parameters comprise high pressure recoup
Scavenge temperature delta is used to determine if engine cavity pressure, P4.8 cavity pressure, engine Speed and
operation is within a predetermined Sigma level. The arbi compressor-discharge (P3 cavity) pressure.
trary Sigma level, for example, can be 3 Sigma, where 3 4. A method as claimed in claim 3 wherein each prede
times 6.8 equals 20.4 deg F. delta, or 11.3 deg C. delta, termined engine parameter is multiplied by an associated
between predicted versus measured value. If the arbitrary coefficient.
Sigma level is Surpassed, for example, to 4Sigma, or 27.2 5. A method as claimed in claim 4 wherein each associ
deg F. (15.1 deg C.), a control trip can be launched. The ated coefficient is a fixed value used to calculate the pre
deltas are taken into consideration at any given Speed or dicted rise value.
power level. 35
6. A method as claimed in claim 1 wherein the step of
While the invention has been described with reference to responding to the comparison value comprises the Step of
a preferred embodiment, it will be understood by those automatically shutting off the engine when the Sigma level
skilled in the art that various changes may be made and is greater than a predetermined allowed Sigma level.
equivalents may be substituted for elements thereof without
departing from the Scope of the invention. In addition, many k k k k k