24

R&D Review of Toyota CRDL Vol. 41 No. 1

Self-acting Air-lubricated Bearing without Oil Lubrication

Masayoshi Otsuka

ResearchReport

Abstract

One of the advanced technologies incorporatedinto a micro gas turbine (MGT) is the dynamicair-lubricated (no oil lubrication) bearing. Weinvestigated the performance characteristics of aself-acting air-lubricated bearing in a 50-kWeMGT. From the results of our experiments, wewere able to clarify the following.

(1) The thrust load of 300 N incurred by a 50-kWe MGT cannot be supported by the dynamicair pressure generated by a pocket-type bearingwith a micro-meter size groove.

(2) The target performance (a film thickness ofat least 20 μm at a load of 300 N) can be achievedby introducing compressed air (air pressure ofequal to or more than 120 kPa) into the rotor discsurface through a pocket-groove.

(3) The performance characteristics of the self-acting air-lubricated bearing were improved whena new structure featuring a rear-mounted thrustbearing with a sheet-spring was adopted.

Keywords Air bearing, Thrust load, Film thickness, Performance evaluation, Micro gas turbine,Cogeneration

Special Issue Core Technology of Micro Gas Turbine for Cogeneration System

R&D Review of Toyota CRDL Vol. 41 No. 1

1. Introduction

For a micro gas turbine (MGT), switching fromconventional oil-lubricated bearings to self-actingair-lubricated bearings (hereafter, "air bearings"),offers the following advantages.

(1) Thermal efficiency is improved because theuse of the air bearing reduces the friction lossincurred by a high-speed rotating shaft.

(2) The size and initial cost of the MGT arereduced because an oil supply device isunnecessary.

(3) Maintenance costs are also reduced becausethere is no oil and no oil supply device.

When an air bearing is adopted for the shaft of anMGT, the specifications should be considered asfollows.

(1) The shaft of the MGT rotates extremely quickly, at up to 80,000 rpm.

(2) In an MGT, the thrust load is greater than theradial load.

(3) Shaft vibration occurs in the event of any shaftimbalance.

(4) Air friction can lead to largerises in the temperature.

If an air bearing is to be appliedto an MGT, however, thefollowing characteristics of thebearing must be improved.

(1) Low stiffness and low loadcapacity.

(2) Frequent occurrence ofunstable oscillation.

(3) Contact between the bearingand the shaft at start-up andstop.

In this study, we investigated thecharacteristics of a thrust airbearing.

There are several kinds of thrustair bearing, as listed in Table 1.The 28-kWe MGT commercializedby Capstone Engineering in thelater half of the 1990s uses alaminated foil bearing, like thatshown in Fig. 1. Also, an MGTthat uses an air bearing with a

bump-type foil, as shown in Fig. 2, was reported byHonda at the 2003 ASME conference.3) Theapplication of foil bearings to an MGT has become anew mainstream technology.

Unfortunately, the complex configuration of foilbearings may make them too expensive forcommercialization. Also, foils cannot support highloads as they are easily deformed. Therefore, wedecided to examine the formation of a pocket-grooveon a rigid, broad bearing surface that would notdeform. The appearance of the pocket-groove typebearing is shown in Fig. 3, and a schematic of theprinciple by which dynamic air pressure is generatedis shown in Fig. 4. We know that the air flow in thecircumferential direction, as induced by theinteraction with the air viscosity and the rotatingsurface, is compressed into the groove, such that thepressure on the surface increases. This air pressurecan support the thrust load that is caused by thepressure difference between the compressor and

25

Step

Step bearing

Grooved

Spiral grooved

Pivoted

Tillting pad

Compliant

Foil

Classification

Typical name

Typical shape

Characteristic

Structure is simple.Processing is easy.Stability is high.Load-index is low.

Structure is simple.Processing is easy.Stability is low,to be fewload-index is high.

Structure is complex.Load-index is high.Precision is easy, to be few.

Structure is complex.Precision is easy, to be few.

Table 1 Types of thrust bearings.1)

Top foil

Rotor disk

DumperBase

Fig. 1 Foil bearing.

Top foil Bump foil

Fig. 2 Bump type foil bearing.

turbine of the MGT.It is necessary to prevent the leakage from the

periphery of the surface as this causes the airpressure to decrease. This leakage is caused by airflowing in the radial direction, as a result of thecentrifugal force. A pocket shape is effective forpreventing this air leakage. Figure 5 shows acomparison of a pocket-groove-W type, a pocket-groove-N type, a step-land type, and a taper-landtype and indicates that the pocket-groove type offersthe best performance characteristics for a given filmthickness.

2. Target air bearing performance for an MGT

2. 1 Estimation of thrust load and film thicknessTo estimate the thrust load, we assumed the use of

a 50-kWe class MGT. As shown in Fig. 6, the thrustload increases in proportion to the rotor speed, butthe film thickness decreases. At the rated rotational

speed of 80,000 rpm, the thrust load is estimated tobe about 260 N. It is assumed that a film thicknessequal to or greater than 22 μm is needed to supportthis thrust load.

In the early stages of our investigation, we aimedtemporarily at a target film thickness equal to orgreater than 20 μm for a thrust load of 300 N at arotor speed of 80,000 rpm.

2. 2 Bearing characteristicsWe calculated a bearing outer diameter that

satisfies the target value by using a relationalexpression for the bearing characteristics.

The following expression links the bearing number

26

R&D Review of Toyota CRDL Vol. 41 No. 1

Number of pad : 6

Electrical discharge machining

Ratio of inner & outer diameter : 0.4

Ratio of groove height : 0.4

Depth of grooved : 20 m

64 mm

Type : Pocket-groove -N

Fig. 3 The shape and the specification of the pocket-groovetype bearing.

Film

thic

knes

s (

m)

Rotational speed ( 1000 rpm)

Taper-land

Step-land

Pocket-groove -N

0 10 20 30 40 50 60 70 80 90 1000

10

20

30

40

5

15

25

35

Taper-land

Step-land

Pocket-groove -W

Pocket-groove -N

Pocket-groove -W

Fig. 5 The comparison between a pocket-groove typeand a step type.

Face of bearing

Rising of pressure

Air flow

Rotor disk

Fig. 4 The principle of the groove-type dynamic airpressure bearing.

Film thickness

Film

thic

knes

s (

m)

200

30 40 50 60 70 80

150

100

50

0

250

300

Thr

ust

load

(N

)

0

10

20

30

40

50

60

Thrust load

Rotational speed ( 1000 rpm)

Fig. 6 The estimation of thrust load.

and the dimensionless load carrying capacity.The bearing Number c is defined by Eq. (1)

• • • • • • • • • • • • • (1)

μ : Viscosityω : Angular velocityPa : Ambient pressureh : Film thickness

The dimensionless load carrying capacity Wn isdefined by Eq. (2)

• • • • • • • • • • • • • • • (2)

W : LoadRo : Bearing outer radiusRi : Bearing inner radiusπ : Circular constant

Figure 7 shows the relationship between thebearing number and the dimensionless load carryingcapacity. Figure 8 shows the relationship betweenthe bearing outer diameter and the load when thefilm thickness is held constant at 20 μm with a ratioof 0.4 between the outer and the inner diameter.From Fig. 8, we find that an outer diameter equal toor more than 100 mm is needed in order to supportloads of 300 N at a rotor speed of 80,000 rpm.

3. Investigation of design of self-acting air-lubricated bearing

We studied the thrust air bearings in two stages.(1) In the first stage, we evaluated the performance

2 2o ia (R -R)P

Wn =W

2 2o i

c 2a

3 (R -R )P h

=

characteristics of the air bearing at rotational speedsup to 80,000 rpm. We chose to do this because therewere no reports on the performance of air bearings atultra high-speed rotational speeds.

(2) In the second stage, we demonstrated the thrustload of 300 N at a rotational speed of 80,000 rpmbecause, again, there were no reports in theliterature.

3. 1 Basic test at high rotational speed3. 1. 1 Test rig and experimental methodWe developed a basic test rig for evaluating the

bearing characteristics. A schematic of this test rigis shown in Figs. 9. A photograph of the rig isshown in Fig. 10. The test rig consists of a rotordisk, a drive unit, a bearing attachment, bearingsupports, a load and associated measurementinstruments, and an air-damped mount.

The features of the test rig are as follows: 1) The rotational shaft is oriented horizontally

even though it is vertical in practical applications.2) A bearing surface is configured on the

stationary side even though it is on the rotating sidein practical applications. With this configuration,the surface of the circular disc can support the loaduniformly, and the load can be adjusted easily. Thisconfiguration also makes it easy to measure the filmthickness, the friction torque, and so.

The specifications of the test rig are as follows.(1) The rotor speed is set to between 5,000 rpm

and 100,000 rpm (varied continuously)(2) The load is increased from 15 N to 55 N (in

0.5-N increments)(3) The bearing outer diameter is 64 mm or 68 mm

(to a maximum of 100 mm)

27

R&D Review of Toyota CRDL Vol. 41 No. 1

Outer diameter of bearings (mm)

0

20

40 60 80 100 120 140 160

800

600

400

200

1000

300

Loa

d (N

) Film thickness of 20 m

Fig. 8 The estimation of the bearing outer diameter.

0

0.1

0.2

0.3

0.4

0.5

0 5 10 15 20 25 30

Dim

ensi

onle

ss lo

ad c

arri

ng c

apac

ity

Wn

Bearing number

Target line of characteristics

Target

Ri / Ro = 0.4

0.5

Ri / Ro = 0.6

0.7

Fig. 7 Characteristics of spiral grooved air bearings.

The measured terms (which are mainly the staticcharacteristics) are as follows.

(1) The film thickness is defined as the bearing clearance, as measured by an eddy current typegap sensor.

Film thickness = displacement of the bearing - displacement of the rotor disk

(2) Friction torque is calculated by measuring witha load meter.

Friction torque = load * radius (measured position)

A bearing armature is lowered to a point near therotor surface once the desired rotor speed has beenreached. Then, we confirm that the bearing

armature is not touching the rotor surface and that itis stable. Once the bearing system is stable, thedisplacement between the bearing and the rotor disk,and the load incurred by the viscous friction of theair and the temperature, are measured. The filmthickness is calculated from the displacement, whilethe friction torque is calculated from the load.

3. 1. 2 Test resultsThe specifications of the test bearings are listed in

Table 2. The film thickness and the friction torque

28

R&D Review of Toyota CRDL Vol. 41 No. 1

Test bearing

Rotor disk

Gap sensor

(Load meter)

LoadAttachment

Friction torque

Film thickness

Hydrostatic airbearing

(15 55N)

(80,000 rpm)

(a) Enlargement on the measuring part.

Attachment Stator

Supportstand

Cover Test bearing

High-frequency motor

Mount with airdamper

Gap sensor

Hydrostatic air bearing

(max 105,000 rpm)

Weight

Weight release structure

Micrometer

Rotor disk

(b) Basic test rig.

Fig. 10 Photograph of the basic test rig.

Outer diameterRatio of outer toinner diameterGroove angle

Pad angle

Ratio of groove height

Groove depthTotal number of padField coarseness

A

View A

hs

height

Table 2 Specification of a thrust air bearing.

Fig. 9 Schematic of the basic test rig.

of the pocket-groove type bearing are shown inFigs. 11. Figure 11(a) shows the obtainedperformance characteristics for a rotor speed from20,000 rpm to 80,000 rpm at a bearing load of 15 N.Figure 11(b) shows the performance characteristicsfor loads between 15 N and 35 N at a rotationalspeed of 80,000 rpm. The film thickness is almostproportional to the rotor speed and is inverselyproportional to the load. The friction torque isproportional to both the rotor speed and the load.Although the film thickness is almost proportional tothe rotor speed up to 40,000 rpm, the film thicknessat rotor speeds in excess of 40,000 rpm increases at alow rate. The friction torque is directly proportionalto the rotor speed. The surface temperature of thebearings rises by about 20 OC between 20,000 and80,000 rpm, which constitutes a considerabledifferential.

Our preliminary investigation revealed themeasured characteristics to be inferior to theestimated target characteristic line, as shown in Fig. 12.Therefore, the performance characteristics of thetrial air bearings must be improved to reach thetarget.

3. 2 Factors leading to non-achievement oftarget performance and measures

In this chapter, we consider the reasons for thefilm thickness being smaller than the target value.

3. 2. 1 Degradation factors(1) Insufficient increase in dynamic air pressure

The dynamic pressure that is generated in thepocket-groove is relatively small, compared to theinitial estimate.

(2) Decrease in the generated dynamic air pressureThe generated dynamic air pressure fell for several

reasons, as follows.(i) Misalignment and deformation of the rotor disk

and the bearing The rotor disc of the thrust bearing and the

bearing's surface with the groove are completelyparallel in the ideal design. However, there isdeformation of the rotor disk caused by geometricdifferences, centrifugal force, and the load. There isalso an inclination of several microns to 10 μmcaused by inaccuracies in the forming accuracy andin the perpendicularity of the axis, and so on.Because this inclination causes a crack in the non-

29

R&D Review of Toyota CRDL Vol. 41 No. 1

Bearing number c

Dim

ensi

onle

ss lo

ad c

arri

ng c

apac

ity W

n

Target line of characteristic

0 5 10 15 200

0.1

0.2

0.3

Experiments

Fig. 12 Difference of the bearing characteristic betweenexperiments and target line.

Film

thic

knes

s (

m)

&fr

ictio

n to

rque

(m

Nm

)

Load (N)

Rotational Speed:80,000 rpm fixed

10 15 20 25 30 35 40

10

15

20

25

30

0

5

Film thickness

Friction torque

Fig. 11(b) Characteristics with load.

Rotational speed (thousands rpm)

Film

thic

knes

s (

m)

&fr

ictio

n to

rque

(m

Nm

)

Atm

osph

ere

tem

pera

ture

&su

rfac

etem

pera

ture

()

Load:15N fixed

0

5

10

15

20

25

30

0 20 40 60 80 1000

10

20

30

40

50

60

Film thickness

Film thickness

Friction torque

Atmosphere temperature

Surface temperature

Friction torque

Fig. 11(a) Characteristics with rotor speed.

parallel gap, the increased air pressure falls relativeto that on the side of the larger crack (Fig. 13(a)).Also, because the rotational speed of 80,000 rpm isvery high and the centrifugal force extends the rotordisc surface in the radial direction, distortion of thecircular disc surface occurs and the flatness of thesurface is lost (Fig. 13(b)).

The values calculated by a FEM analysis areshown in Fig. 14. This calculation indicates thatpart of the rotor disk tip is bent as a result of thegenerated pressure.

(ii) Centrifugal whirling of the rotor disk causedby eccentricity of the rotational axis

Centrifugal whirling occurs when there is adifference in the straightness of the rotor axis.Because of the misalignment and the deformation ofthe disk and bearing surfaces, the dynamic airpressure decreases (Fig. 13(c)).

3. 2. 2 ImprovementsThe following improvements were reviewed as a

means of overcoming those factors that lead todegradation.

(1) Optimization of the groove shapeThe bearing characteristics are influenced by the

shape of the bearing surface. The results of ourexamining the groove depth and the groove heightare shown in Figs. 15(a) and (b)(also see Table 2).These results indicate that the bearing characteristicscan be improved by making the groove deeper andexpanding it in the radial direction. Figure 15(c)shows that the bearing characteristics approach thetarget line as a result of changing the bearing grooveshape.

(2) Air introductionIt is possible to improve the bearing

characteristics when the static pressure rises,based on the estimation using the bearingcharacteristic equations. We confirmed thatthe film thickness expanded as a result ofintroducing compressed air into the rotor discsurface from the centre of the test rig. Figure16 shows the groove bearing characteristicimprovements obtained by introducing thecompressed air. The film thickness expandsas a result of increasing the air pressure.

(3) Shape and surface roughness of therotor disk

We created a convex area in the centre of the rotoron the bearing surface, and improved the surfaceprocessing precision. As a result of this, we shouldaim to reduce the flexing caused by the centrifugalforce and the surface friction.

(4) Provision of sheet-spring structure supportWe employed a structure supported with a sheet-

spring to follow the movement of the armature. Avariety of spring-based structures have beenproposed in patents, but we contrived the structuredescribed below. A lamina disk with a thickness of0.1 to 0.3 mm and a surface groove is positionedbehind the bearings, the perimeter of the disk isfixed to the bearing, and the internal circumferenceof the disk is fixed to the bearing base. A bearingwith this structure can decline even if the pressureload is partially taken by the perimeter of the disk. Atest result is shown in Fig. 17. The film thickness

30

R&D Review of Toyota CRDL Vol. 41 No. 1

Bending

0

-5

-10

-15

-200 10 20 30 40 50

Def

orm

atio

n (

m

)

Radius R (mm)

N = 80,000 rpm

W = 200 N

R = 500

Fig. 14 Deformation of the rotor disk calculated by a FEM analysis.

Film thickness

Rotor disk Bearing

The circumference of thedeflection of a disk

Time Qua

ntity

of

disp

lace

men

t

80,000 rpm, 30 kgf

(a)

(b)(c)

20 m

Fig. 13 Dynamic behavior of the rotor disk and thebearing.

becomes large as a result of employing thesheet-spring structure, so we can conclude thatthis structure is very effective.

4. Proof test

4. 1 High-speed and high-load testThe purpose of this test was to achieve the

target value (rotor speed of 80,000 rpm, load of300 N, film thickness in excess of 20 μm) whilereducing the bearing outer diameter. Whenincorporating a thrust bearing into an actualMGT, it is important that the bearing outerdiameter be made as small as possible, becausean outer diameter of 100 mm is too large for the50-kWe class MGT.

31

R&D Review of Toyota CRDL Vol. 41 No. 1

Target line of characteristic

0 5 10 15 20c

0

0.1

0.2

0.3

Dim

ensi

onle

ss lo

ad c

arri

ng c

apac

ity W

n

Bearing number

Fig. 15(c) Bearing characteristics.

0

5

10

15

20

0 0.2 0.4 0.6 0.80

10

20

30

40

0 0.2 0.4 0.6 0.8

Film

thic

knes

s (

m

)

Ratio of the groove height

(a) Film thickness

Load:15N fixed

Fri

ctio

n to

rque

(m

Nm

)

(b) Friction torque

Rotational Speed:80,000rpm fixed

Ratio of the groove height1 1

80,000 rpm, 15N80,000 rpm, 35N

20,000 rpm, 15N80,000 rpm, 15N80,000 rpm, 35N

20,000 rpm, 15N

Fig. 15(b) Influence of the groove height.

0

10

20

30

40

0 20 40 60 80Depth ( m )

Film

thic

knes

s (

m )

Load:15N fixed

Depth ( m )

(a) Film thickness (b) Friction torque

Rotational speed:80,000 rpm fixed

Fric

tion

torq

um (

mN

m )

20,000 rpm, 15N80,000 rpm, 15N80,000 rpm, 35N

20

15

10

5

00 20 40 60 80

20,000 rpm, 15N

80,000 rpm, 15N80,000 rpm, 35N

Fig. 15(a) Influence of the groove depth.

4. 1. 1 Test rigOur high-load test rig is shown in Fig. 18.

The rotational shaft of the test rig isoriented horizontally, unlike the basic testrig in which the shaft is oriented vertically.The structure of the bearings is basicallysame as in the basic test rig, but differs inthat the load is supported by a coil springand the rotor is driven by a turbine.

4. 1. 2 Test resultsAs the load was increased, the rotor disk

would tend to contact the bearings moreeasily because the film thickness decreasesas the load increases. Therefore, the testwas carried out after setting the rotor speedto 80,000 rpm and increasing the filmthickness by introducing compressed airfrom the start of the test. The resultsobtained for a bearing outer diameter of 100mm are shown in Fig. 19. The filmthickness remained relatively constant atabout 70 μm for loads of 50 N to 250 N,because the inlet air pressure was increasedfrom 30 kPa to 100 kPa. After that, the airpressure was held at 100 kPa and the loadwas increased from 250 N to 300 N. Then,the film thickness decreased, but the filmthickness of 55 μm at 300 N surpassed thetarget value of at least 20 μm. The ambienttemperature increased by 120 OC becausethe load of 300 N had an exothermic effect.

Figure 20 shows the results of examiningthe effect of varying the supplied airpressure for bearing outer diameters of 100mm and 80 mm. The film thickness fell to

32

R&D Review of Toyota CRDL Vol. 41 No. 1

HandleCoil spring

Load cell

Stator

Hydrostatic air bearing

Cover Attachment

Test bearing

Rotor disk

The turbine for a drive

Compressure test rig

Load cell

Fig. 18 Test rig for high-load experiment.

Bearing number c

Dim

ensi

onle

ss lo

ad c

arri

ng c

apac

ity W

n

Target line of characteristic

Entrance air pressure (kPa)

Film

thic

knes

s (

m

)

Air

Load characteristic without air introduction

80,000 rpm,35N

Air introduction

10 20 30 40 50

80,000 rpm,35N10

20

30

40

50

00

10 205 1500.0

0.1

0.15

0.2

0.05

0.25

Air

Fig. 16 Characteristics of the groove bearing improved by the airintroduction.

Surface ofthe bearing

Entrance supply air pressure (kPa)

Film

thic

knes

s (

m

)

Bearing Attachment

Sheet spring

Without bord spring

100 110 120 130 140 150

rotor

t = 0.2mm

t = 0.15mm

0

20

40

60

80

100

Fig. 17 Examination of spring structure. Film thickness increased byair introduction and the effect of the structure of the bearingwith a sheet spring.

30 μm when the air pressure was dropped to 80 kPafor a bearing outer diameter of 100 mm. Becausethe performance declines in case of a bearing outerdiameter of 80 mm, the inlet air pressure must beincreased in order to achieve the same film thicknessas that obtained for a diameter of 100 mm. Thefigure shows that the target of a film thickness of atleast 20 μm can be secured when the air pressure ismaintained at least 120 kPa. In other words, we cansay that an air bearing for an MGT cannot berealized successfully without introducing air.Actually, an MGT can use air fromits compressor. Introducingcompressed air from the impellerexit would allow us to attain the goalneeded for an MGT. Because thetarget MGT has an impeller deliverypressure of about 150 kPa, we couldrealize an air bearing with an outerdiameter of 80 mm.

4. 2 Proof test in an actual rotorWe incorporated the developed

thrust air bearings into a compressortest rig which closely approximatedthe shape of an actual rotor, and thenevaluated the performancecharacteristics.

We investigated the followingitems.

(1) Handling of rotor resonancecaused by imbalance, particularlyat start-up and stop

(2) Effect of the performance

improvement attained through impeller deliverypressure introduction

Regarding the resonance caused by imbalance ofthe rotor assembly, our computational analysisindicated that the 1st and the 2nd rotational modesexist in the low-speed range of around 6,000 rpmand 20,000 rpm, as shown in Fig. 21. Theseresonance points of the rotor dynamics must beovercome. The resonant frequency of the rotorassembly is not a problem because a bending modeof the rotor exists at rotational speeds equal to orgreater than 85,000 rpm, which is in excess of the

33

R&D Review of Toyota CRDL Vol. 41 No. 1

Rotational speed = 80,000 rpm ; Load = 300 N

Entrance supplied air pressure (kPa)

Film

thic

knes

s (

m

)

100

40200

20

40

60

80

20018016014012010080600

Fig. 20 The difference of the film thickness between80mm and 100mm of the bearing outer diameter.

Rotational speed ( 10,000rpm)

Rigid mode

Bend mode

0 1 2 3 4 5 6 7 8 9 100

10

20

30

40

50

Am

plitu

de (

m)

Position no.1

Position no.13

Position no.18

Fig. 21 Resonant frequency of the shaft assembly.

Film thickness

Friction torque

Atmosphere temperature

Supplied air pressure

0 50 100 150 200 250 300

Load (N)

0

20

40

60

80

100

120

0

20

40

60

80

100

120

Sup

plie

d ai

r pr

essu

re (

kPa)

Film

thic

knes

s (

m)

Fric

tion

torq

ue (

mN

m)

A

tmos

pher

e te

mpe

ratu

re (

Outer diameterRatio of outer toinner diameterGroove angle

Pad angle

Ratio of groove height

Groove depthTotal number of padField coarseness

Fig. 19 Experimental result.

rated speed of the target MGT.To cope with the contact between

the bearings and the rotor disk atstart-up and stop, some finishes forthe bearing surface were reviewedand, as a result, we selected thefollowing finish.

Bearing surface: The material washeat-treated (quenching, tempering)and the strip surface roughness wasminimized in the manufacturingprocess.

Armature surface: The surface wascoated with solid MoS2 lubricant.

4. 2. 1 Groove shape on bearingsurface

Considering the actual size of theMGT, we decided to employ anoutside bearing diameter of 80 mm,and the bearing surface groove shapeshown in Fig. 22. Given our previousresults, the groove shape of the pockettype bearing surface was designed asshown in the figure and the groovewas kept shallow at the outercircumference.

4. 2. 2 Assembly test rigA schematic of our assembly test

rig is shown in Fig. 23. The test rigwas remodeled from one used for acentrifugal compressor whichemployed ball bearings. The bearingpart of the test rig is such that thethrust load is carried by the airbearing but the radial load is carriedby an oil-lubricated roller bearing. Inthe first part of this test, the operationof the air bearings was confirmedwhen sufficient compressed air wassupplied. In the next step, weconfirmed that sufficient compressedair could be delivered from theimpeller exit.

4. 2. 3 ResultsThe results of introducing air from

the impeller exit are shown in Fig. 24.These results indicate a time span

34

R&D Review of Toyota CRDL Vol. 41 No. 1

Solid-lubricant coatingType : Pocket-groove-W

Outer diameter : 80 mm Groove depth : 60 m Thickness of sheet spring : 0.15 mm

without surface coating

Rotor disk Diameter thickness : 80 5 mm Material : Titanium alloy Surface : Solid-lubricant coatin

Shallow grooveDeep groove

Surface finish : 0.6z

Bearing Rotor disk

Bearing

Fig. 22 The shape of the air bearing.

Roller bearing(with oil lubrication)

Rotor disk

tImpeller

Diffuser

Gap sensor

Shaft

Thrust air bearing

Air

External compressed air supplyInternal compressed air supply

Fig. 23 Schematic of the assembly test rig.

Displacement of disk

Thrust load (estimated)

Supplied air pressure

Displacement of shaft end

Rotor speed

70 170 270 370 470 570 670 770 870Time (sec)

-300

-240

-180

-120

-60

0

60

120

180

240

300

Dis

plac

emen

t (m

)

-10

0

10

20

30

40

50

60

70

80

Rot

atio

nal s

peed

(10

00rp

m),

air

pre

ssur

e (k

Pa),

load

(0.

1N)

N =

Fig. 24 Experimental result (1).

from start-up to the point where a rotational speed of60,000 rpm is reached. We were able to confirmstable operation of the air bearings. The impellerdelivery pressure increased depending on theincrease in the rotor speed, while the displacementof the rotor disk fell. The pressure delivered by theimpeller was 80 kPa at 60,000 rpm, while weestimated a thrust load of 300 N. The impeller-delivered pressure was lower because the rotor speedwas lower than that of the high-load test, but thethrust load was comparatively high. The thrust loadactually improved slightly compared with the valuewhich was calculated from the impeller-deliveredpressure.

Also, no problems were observed in continuousoperation equal to or more than 15 minutes at 50,000rpm, as shown in Fig. 25.

5. Summary

A dynamic air pressure bearing without oillubrication is an important core technology for amicro gas turbine (MGT). We investigated theperformance characteristics of a self-acting air-lubricated bearing for a 50-kWe class MGT. As anair bearing for the MGT, we investigated the effectof different groove types on dynamic air pressurebearings. Our investigation of the thrust bearingclarified the following points.

(1) In the early stages of our investigation, weaimed at a target film thickness of at least 20 μm fora thrust load of 300 N and a rotor speed of 80,000rpm.

(2) Pocket-type groove shapes were reviewedusing a basic test rig running at speeds between20,000 and 80,000 rpm, and we selected the best ofthese shapes.

(3) When employing a high-load test rig capableof applying a load of 300 N at a rotational speed of80,000 rpm, we were able to achieve the target value(more than 20μm) by adapting a sheet-springstructure with a disc diameter of 80 mm, less thanthe 100 mm of the preliminary design, and byintroducing air at a pressure equal to or more than120 kPa.

(4) The developed air bearing was evaluated at amaximum rotor speed of 60,000 rpm with anassembly test rig which imitated the actual gasgenerator of an MGT, and the test involvingcontinuous running for 15 minutes at 50,000 rpmwas carried out. As a result, we were able to clarifythe practicality of using a pocket-groove typedynamic air pressure bearing supported by a sheet-spring.

References

1) Togo, S. : "Kitaijikuuke -Sekkei kara Seisaku made-",(in Japanese), (1999), Kyoritsu Shuppan

2) RPI-MTI Gas-Bearing Design Course3) Microturbine Equipment Panel, ASME Turbo Expo.,

Atlanta, 16-19 June 20034) Shimura, K. :"Microturbine Research at Honda",

ASME Pap. No.2003-GT-39055(2003)5) Ochiai, M. and Hashimoto, H. : "Static and Dynamic

Characteristics of High-Speed, Stepped Thrust Gas-Film Bearings", Trans. of JSME, C, (in Japanese),64-628(1998), 305-312

6) Ochiai, M. and Hashimoto, H. : "Measurements ofCompressibility Effects in Stepped Thrust Gas FilmBearing", Tribo. Trans., 42-4(1999), 723-730

7) Hashimoto, H., et al. : 2001nendo Nenjitaikai KouenRonbunshuu, (in Japanese), (2001), JSME

8) Hashimoto, H., et al. : 2002nendo Nenjitaikai KouenRonbunshuu, (in Japanese), (2002), JSME

9) Capstone Turbine co. Ltd. : Jpn. Transl. PCP.H09-510522, (in Japanese)

(Report received on Dec. 23, 2005)

35

R&D Review of Toyota CRDL Vol. 41 No. 1

Masayoshi OtsukaResearch fields : Gas Turbine

(Aerodynamics, CFD Analysis)Academic society : Jpn. Soc. Mech. Eng.,

Gas Turbine Soc. Jpn.

About 15 minutes

Displacement of disk

Thrust load(estimated)

Supplied air pressure

Displacement of shaft end

Rotor speed

0 5 10 15 20Time (min)

60

40

80

-20

-40

0

Dis

plac

emen

t (m

)

20

Fig. 25 Experimental result (2).