AM-MPEX

FOIL BEARING DESIGN MANUAL Prepared By A. Eshel L. Licht

DD C',

Ampex Corporation Research and Advanced Technology Division

RR 71-18
NATIONAL TECHNICAL INFORMATION SERVICE 22151
,ponr' eld. Va

itpdce

by

30 September 1971

Unclassified
Security Classification

DOCUMENT CONTROL DATA - R&D

I
(Security claesfication o Wi4., body of absrect and ld*zXljn mrratation must be datered wfen the overall report le Celaillod) 2. REPORT SECURITY C LASSIICATION oqIGINATIN 0 ACTIVITY (Corporaeo au tr)

Ampex CorporationI 401 Broadway

-Redwood City. California
3 REPORT TITLE

Uncl.
zb. su,

Foil Bearing Design Manual

4. DESCRIPTIVE NOTES (Typ of rpor and incluslv dates)

Technical Report,
5 AUTHOR(S) (Last name. flirt na.e, initial)

Eshel, A., Gross, W.A., Licht, L., Szego, P.

6. REPO AT

DATE

7a. TOTAL NO, OF PAC=

, Hb OF RIF$ O.

30 September 1971

76
*., ORIGINATOR'S REPORT NUMER(S)

aa. CONTRACT OR GRANT NO,

N00014-71-C-0001
b. PROJECT NO.

RR 71-18
__
Sb. OTHER N PORT NO(S) (Any other number tehar may be a@seeined

NR06 2-297
C.

hle ,port)

d.
10. A VA IL ABILITY/LIMITATION NOTICES

This document has been approved for public release; its distribution is unlimited.
Reproduction in whole or in part is permitted for any purpose of the United States
SUPPLEMENTARY NOTS J12. SPONSORING MILITARY ACTIVITY

Government.
11

13. ABSTRACT

Director, Fluid Dynamics Programs, Mathematical and Information Sciences Div. ,Office of Naval Research,Dept. of the Navy. Arlington, Virginia 2217.

A compendium of theoretical and experimental results, available at the present in the field of foil bearings, is presented. Design formulae and ' rief explanations of concepts are given for self-acting and externally pressurized foil-bearings, with reference to applications in the transport of flexible media and the support of rotating machinery.

DD

, 1473 '-A

Unclassified
Security Classification

Uncla ssified
Security Classification =
t4 KEY WORDS
VEY WROLZ WT

:LINK

ALINK
ROLE

B L~KaLN WY

LINK C
ROLE WT

Self Acting Foil Bearings Externally Pressurized Foil Bearings Foil-Bearings for Rotating Machinery

INSTRUCTIONS I. ORIGINATING ACTIVITY: Enter the name and address 10. AVAILABILITY/LIMITATION NOTICES: Enter any lmof the contractor, subcontractor, grantee, Department of Deitations on further dissiminstion of the report. oler than those fense activity o other organization (corporate author) Issuing imposed by security classification, using standard statements the report. such as: 2&. REPORT SECUI TY CLASSIFICATION: Enter the over. (1) "Qualified requesters may obtain copies of this all security classification of the report. Indicate whether report from DDC. "Restricted Data" is included. Marking is to be In accordance with appropriate security regulations. (2) "Foreign announcement and dissemination of this report by DDC is~not authorized." 2b. GROUP: Automatic downgrading is specified in DoD D.1rective 5200. 10 and Armed Forces Industrial Manual. Enter (3) "U. S. Government agencies may obtain copies of the group number. Also, when applicable, show that optional this report directly from DOC. Other qualified DDC markings have been used for Group 3 and Group 4 as authorusers shall request through ized. 3. REPORT TITLE: Enter the complete report title in all (4) "U. S. military agencies may obtain copies of this capital letters. Titles in all cases should be unclassified, report directly from DDC. Other qualified users If a meaningful title cannot be selected without classificeshall request through tion. shlow title classification in all capitals in parenthesis
immediately following the title....__
_

4. DESCRIPTIVE NOTES If appropriate, enter the type of report, e.g., interim, progress, summary, annual, or final. Give the inclusive dates when a specific reporting period is
covered.

(5)

"All distribution of this report is controlled, Qua l ified DDC users shall request through

o. AUTHOR(S) Enter the name(s) of author(s) as shown on . or inthe report. Enter lost name, first name, middle Initial. If military, show rank and branch of service The name of the principal author is an absolute minimum requirement. 6. REPORT DATE: Enter the date of the report as day, month, year, or month, year. If more than one date appears on the report, use date of publication 7a. TOTAL NUMBER OF PAGES: The total page count should follow normal pagination procedures, Le., enter the number of pages containing information, 7b. NUMBER OF REFERENCES: Enter the total number of references cited in the report. 8. CONTRACT OR GRANT NUMBER: If appropriate, enter the applicable number of the contract or grant under which the report was written, Sb, 8c, & 8d. PROJECT NUMBER: Enter the appropriate military department Identification, such as project number, subproject number, system numbers, task number, etc. 9s. ORIGINATOR'S REPORT NUMBER(S): Enter the official report number by which the document will be identified and controlled by the originating activity. This number must be unique to this report. 9b. OTHER REPORT NUMBER(S): If the report has been assigned any other report numbers (either by the originator or by the sponsor), also enter this number(s).

If the report has been furnished to the Office of Technical Services, Department of Commerce, for sale to the public, indicate this fact and enter the price, If known. IL SUPPLEMENTARY NOTES: Use for additional explanatory notes. 12. SPONSORING MILITARY ACTIVITY: Enter the name of the departmental project office or laboratory sponsoring (paying for) the research and development. include address. 13. ABSTRACT: Enter an abstract giving a brief and factual summary of the document indicative of the report, even though it may also appear elsewhere in the body of the technical report. If additional space is required, a continuation sheet shall be attached. It is highly desirable that the abstract of classified reports be unclassified. Each paragraph of the abstra't shall end with an indication of the military security classification of the information in the paragraph, represented as ('S), (S), (C), or (U). There is no limitation on the length of the abstract. However, the suggested length is from 15C to 225 words. 14. KEY WORDS: Key words are technically meaningful terms or short phrases that characterize a report and may be used as index entries for cataloging the report. Key words must be selected so that no security classification is required. Idenfiers. such as equipment model designation, trade name, nilitary project codename, geographic location, may be used as key words but will be followed by an indication of technical context. The assignment of links, rules, and weights Is optional.

Unclassified
Security Classification

AMPEX

FOIL BEARING DESIGN MANUAL

Prepared under:

Contract No. Nonr-N00014-71-C-0001 30 September 1971 Department of Defense Atomic Energy Commission National Aeronautics and Space Administration

Supported jointly by:

Administered by:

S. Doroff Fluid Dynamics Branch Office of Naval Research Department of the Navy

Prepared by:

"

A. Eshel and L. Licht Members of the Research Staff

Approved by: P . SzegA Manager, Mechanics S!ction

Approved by z-

.- --

W. A. Gross Vice President, Research & Advanced Technology Director of Research

This document has been approved 1or pub.i- release; Its dlstributlO., z unlrt!ttd. Reproduction In whole or in part Is permitted for Any |urp's,.e of the United Sta'ez
Govern,ent.

RR 71-18

iii

AMPEX

ACKNOWLEDGEMENTS

The authors wish to express their gratitude for the useful suggestions received from Dr. W. A. Gross, Mr. P. Szego and Mr. M. Wildmann of the Ampex Corporation in the preparation of this manual. Mr. S. Doroff of the Office of Naval Research provided vigorous support to this activity.

RR 71-18

AMPEX

ABSTRACT

A compendium of theoretical and experimental results, available at the present in the field of foil bearings, is presented. Design formulae and brief explanations of concepts are given for self-acting and externally pressurized fcil-bearings, with reference to applications in the transport of flexible media and the support of rotating machinery.

RR 71-18

vii

Preceding page blank

AMPEX

PREFACE

Considerable effort has been expended in the past few years on the study of foil bearings. The purpose of this manual is to extract from the numerous papers and reports the information which is relevant to foil bearing design and to organize it in a form which may be easily accessed and updated. The terminology and basic physical facts are explained in the
Introduction. The user of this manual should acquaint himself with the contents of the Introduction, so that he may easily refer to the particular information required. To make each chapter self contained, each chapter is preceded by nomenclature and followed by a list of references particular to the chap-

ter. References are designated by numbers in brackets.

RR 71-18

viii

_-

AMPEX

CONTENTS Page 1.0 INTRODUCTION 1.1 Definitions and Classification 1. 2 Applications 1.3 Foil Bearings vs. Rigid-Surface Bearings 2.0 SELF ACTING FOIL BEARINGS 2.1 Steady State Characteristics 2. 1. 1 Determination of Steady State Gap in the Uniformity Region 2.1.2 Determination of the Steady State ExitUndulations in the Exit Region 2.1.3 Remarks on the Steady State Characteristics . 2-16 2-5 1-3 1-3 1-3 1-4 2-5 2-5

of Finite Width Foil Bearings

2.1.4 Film Thickness of Foil Bearings of Finite Length (Fixed Points of Attachment) 2.2 Response to Step Disturbances 2.3 Response to Periodic Excitation 3.0 EXTERNALLY PRESSURIZED FOIL BEARINGS 3.1 Steady State Characteristics 3. . 1 Interaction of Self-Acting and Pressurization Effects 3.2 Response to Step Disturbances 3.3 Response to Moving Pressure Sources

2-16

2-22 2-26 2-27 3-3 3-3

3-3 3-7 3-7

RR 71-18

ix

AMPEX

CONTENTS (con't) page 4.0 FOIL-BEARINGS FOR ROTATING MACHINERY 4.1 Advantages of Foil Bearings 4.2 Steady State Characteristics 4.3 Stiffness and Damping Coefficients 4.4 Design Considerations Related to the Bearing Frequency Response 4.5 Example of a Turbo Alternator Bearing Design 4-8 4-15 4-5 4-5 4-5 4-3

RR 71-18

AMPEX

NO MENCLATURE (Chapter 1)

b e h hmi n h T W 6 0

Half-width of foil Eccentricity of rigid-surface bearing Fluid film thickness Minimum film thickness Nominal foil-bearing film-thickness tension per unit width of foil Tension per unit width of foil Bearing load Width of non-uniform edge zone of foil bearing Wrap angle

RR 71-18

1-1

AMPEX--

RR

17118

-AMPEX

1. 0 INTRODUCTION

1. 1

Definitions and Classification Flexible webs of paper, plastic and metal under tension, trans-

ported over cylindrical rollers and guides, can be supported on thin films of air or liquid. foil. Conversely, a high-speed rotor can be floated on a fluid film entrained between the journal and the surface of a stretched flexible The term foil bearing applies to both types of support [ 1. 1, 1. 2). A foil bearing therefore, consists of two surfaces, one sensibly rigid anu the other relatively thin and flexible, separated by a fluid film. The foil bearing is capable of supporting a load by virtue of pressure induced in the lubricating film through the relative motion of surfaces, or by means of an external pressure supply. pressurized foil bearings. 1.2 Applications In the manufacture and processing of foil, plastic film and paper, the foil bearing is present, though perhaps unrecognized, at the multiplicity of rolls, drums and guides. Its function and purpose may involve minimization In analogy with rigid-surface, fluidfilm bearings, we thus distinguish between self-acting and externally-

of friction, elimination of roll inertia and protection of surface finish and fresh coatings. In magnetic recording, instead of maximizing the web-to-guide separation, the problem is frequently that of reducing, or of completely eliminating the air gap between the tape and the recording head.

RR 71-18

1-3

Preceding page blank

AMPEX

The trend toward high-speed, high-temperature turbomachines

dictates the use of uinconventional lubricants and bearings, such as gaslubricated bearings. here, the foil bearing presents itself as one of the few, feasible alternatives, because of its attractive dynamic, thermal and wear characteristics.

1.3

Foil Bearings versus Rigid Surface Bearings

The most important characteristics of foil bearings and the

essential differences in the behavior of foil bearings and rigid-surface bearings are a consequence of the effect of flexibility on the shape of the lubricating film. A qualitative comparison of film shapes in a rigid, partialarc journal bearing and in a foil bearing is made in Fig. 1. 2. 1. In the rigid bearing, Fig. 1.2.1a, the lubricating- film consists of a converging region (1), in which the pressure is above ambient, and a diverging region (2), in the major part of which the pressure is subambient. * In the foil bearing, the lubricating film consists of a converging entrance

regio (1), in which the pressure -ises above ambient, an exit region (2), in
which the foil diverges from the journal following a series of undulations and a central region (3), in which both the clearance h and the pressure p remain virtually constant. It will be noted that in addition to the entrance and exit regions, additional transition zones exist along the lateral edges of the foil. The latter are a consequence of the combined effect of rapid decrease of pressure .,

*In liquid-lubricated bearings, subambient pressures cause vaporization and induce air bubbles, so that the film breaks up. In gas-lubricated bearings, subambient pressures are important in the determination of bearing load capacity. **The central region is also termed the uniformity zone.

RR 71-!8

1-4

AMPIX

TT5

4J

-~

ot
.-4 CJr

to

/3 '01

.c

RR 71-18

AMPEX

along the edges and of the anticlastic edge-undulation associated with the The net result is the bending of thin plates into cylindrical shapes tendency of the foil to close the gap along the edges and to inhibit side leakage. Due to this phenomenon, it is possible to predict the performance of foil bearings quite accurately from formulae based on two dimensional (planar) flow.

RR 71-18

1-6

-AMPEX

REFERENCES

1.1

H. Blok and J.J. Van Rossum, "The Foil Bearing - A New Departure in Hydrodynamic Lubrication," Lub Eng., Vol. 9, No. 6, Dec. 1953, 316-.320.

1.2

W. A. Gross, Gas Lubrication, John Wiley and Sons, New York, 1962, 138-141.

1.3
V

Y. C. Fung and W. H. Wittrick "A Boundary Layer Penomenon in the Large Deflection of Thin Plates". App. Math. Vol. 8, p. 191, 1955. Quart. J. Mech. and

RR 71-18

1-7

AMPEX

RR 71-18

1-8

AMPEX

NOMENCLATURE

(Chapter 2)

A (A,) b

Amplitude factor (Eq. Foil half width

2.3.2)

C
D

Compressibility parameter p ro/T ao0

Bending rigidity of foil per unit width; for foil of homogeneous
2 cross section D = Et3/[ 12(l-v )

Young modulus

f
h h*

Frequency
Local fluid film thickness Clearance in uniformity region. maximum pressure. Film thickness at point of

h h
5

Film thickness along centerline Steady-state film thickness Dimensionless clearance H = h/(r
0

H H* k C
0

2/3) * * h!(r o 2/3) 2

Dimensionless clearance in uniformity region H Bearing stiffness per unit width Initial foil length (at tension T) Inertia parameter
0

I p Pa

1/2 paU2 /(T/r

Local film pressure, absolute Atmospheric pressure

RR 71-18

2-1

Preceding page blank

AMPEX

u
r

Speed parameter

Pu

U -FiFo
-U

Radius parameter

_o

P1//Pa P'a
rro

Length parameter

E 4 2pa/P a

Et

T
PT

Tension parameter

PT

Dimensibnless damping per

unit width

c/1

Dimensionless gap

Ph

Pf

Dimensionless frequency

Pf =

"f/Pa

Pk

Dimensionless stiffness per Pk = k/pa unit width

Frequency ratio

P = D

r /U

Stiffness Parameter, S =

T2 2/3
0

t T

Foil thickness (unless otherwise specified), also time Tension per unit width of foil

RR 71-18

2-2

AMPEX

Steady state tension per unit width Preload (initial) tension per unit width

T0

rI r0 U V p x x1
y

Radius of curvature (Fig. 2.15) Radius of foil bearing journal (guide) Algebraic sum of Journal and foil surface velocities Speed of propagation of disturbance Lateral distance from edge of foil Lateral distance from foil centerline
Component of displacement of rotor center

Edge effect number 8

(4D/r 26)1/4

Y(e, w) Phase between tension and gap perturbation (Eq. 2.3.2)

8 Extensional rigidity of foil. section 6 = Et
Ar A Corner angle at inlet of magnetic head (Fig. 2.15) Peak to trough angular distance of last exit undulation Peak to trough normalized angular distance A = 2/3

For foil of homogeneous cross

e9

A
6 It
E

Increment in foil length due to reasons other than stretching. Foil bearing number E = 6 pU/T

Dimensionless lateral distance from edge of foil

r0

e
B

Wrap Anqle Angular position

RR 71-18

2-3

AMPEX

Molecular mean free path Dimensionless molecular mean free path = 2/3

A
1. v

Viscosity of lubricant Poisson's ratio Normalized angular position

=

1/3

Density of ambient f.uid

Dimensionless time

-C =

Ut
2r
0

-1/3 -1/e

(t =time)

0 o

Dimensionless amplitude of tension perturbation. circular frequency of excitation

RR 71-18

2-4

AMPEX

2.0 SELF ACTING FOIL BEARINGS

2.1 2. 1. 1

Steady State Characteristics Determination of Steady State Gap in the Uniformity Region

The gap width in the region of uniformity (Fig. 1.2. lb) is given by the expression
* *

2/3

H rE

(2.1.1)

in which the foil bearing number c order of Y 10.

6 piU/T must be small, at most the 0. 643 3%, provided the following ref-

The constant H

erence conditions C2.1-2.9

are satisfied: S
=

Stiffness

T r0 2 2/3 1 Compressibility C p T
r

<0.8

(2.1.2) (2.1.2)

< 0. 12(213 2

C / Pa Inertia
=

2
<0.05 (2.1.4)
o

T/r

Wrap Angle

1/3

> 6

(2.1.5)

*In this equation, U = IU+

r I according to whether the . surface velocities of the oand of the roll are in the same (sum) or in

U I or U = SI r

opposite (difference) directions.

RR 71-18

2-5

AMPEX

Mean free path

X 2/3 rO
0

< 0.005

(2.1.6)

Foil Width
(where t rep'esents

<<< 2b

(2.1.7)

foil thickness)

Continuity of slope

A0
r r0

=o

(2.1.8)

Constancy vf curvature

= 1

(2.1.9)

The effect of deviations from these restrictions is shown in Figs. 2. 1.1 2. 1. 7. The parameters not explicitly referred to in these figures are assumed to satisfy the above reference requirements. fication, where available, is indicated on the graphs. Experimental veri-

RR 71-18

2-6

AMPEX

_1_5

00

CL

C-4-

.44

T
f
:

-0

~~1

I
4t4 4-4 .4-4

100

C13D

RR2

711

AMPEX

I !
0.5

Udulating

Exit Re ion

Monotonic Exit Region

0.4

"

-r aar

h*
0. 7

) H *r ( - -11 1 1 1 1
2'.6

,
.. I2-8
!

Theory

~~~~Experiment

o [2.4

I"

'

0.1

0.01

0.1

i.

10.

100.

Compressibility Parameter I/C

Fig. 2.1.2 Effect of the Compressibility Parameter 1/C on the Magnitude

of H*

RR 71-1 8

2-8

AMPEX

1.6

1.2 1 4-pau2/(T/ro) H h=H*ro 6M 2/3 U 0.8

0.1

0.2 0.3 0.4 INERTIA PARAMETER I

0.5

Fig. 2. 1 . 3 Effect of the Inertia Parameter I and the Compressibility Parameter C on H* 'L2.9]

RlR 71-1

2-9

AMPEX

Ic

00

C-

00

4~J

Loo
-

:L

I4c

7LU

RR

7148

-1

AMPEX

N~N

00
NN

,7ND

05

RR

71-8

2-1

AMPEX

(a

C4

RR 71-8

2-1

AMPEX--__

2.

07 0
4

0o0r

LD

0)

05

0)

RR0 711

2N 13

------

AMPEX

Example: Find the air film in the region of uniformity of a magnetic head (Fig. 2.1.5) under the following conditions: 0.265 x 10 . 8 X U Pa P T 2.5 piinch = 200 ips = 1.15 x 10
= =

Lbf. sec/in 2

-air

(air)

Lbf. sec2/in

14.7 psi 0.5 Lb/in

e 0

= 100 = 0.25 in.

b E t

0.5in. 5 x 10 5 psi (Mylar) 1 mil (foil-thickness)

S=0.3
Results:

(Mylar)

= 6 U x 0.265 x 10 = 6 T r
_

x 200

=6. 35 x 10 -

0.5 0

0.25

Et 3 2

2 2/3

5 x 105 x (10 3 ) 3

62/3

12(1-v )Tr

12(1-0.3 )0.5x0.25 (6.35 x10

RR 71-18

2-14

AMPEX

Aey
1/3 /
-

5 xIr 180
(6.35 x 10 61/3 T 0.5

4.73

1 C

=00085 <0.12

pa/r 1/2
=

14.7/0.25
U2

1/2 x 1.15 x 10 0.5/0.25

x2023
115 x 10 <0.05

P7 11.x200

T/rO

1/3

(6.35x 10

= 9.45 > 6 0 xqr/180 -61/3

-2/3=

-2.5x

10 6 (6. 35x 106)2/3 =2.88x 102

o /0.001 x 0.25 .=
0

0.25
0.0158 << 0.5

The reference conditions (2.1.2), (2.1.6), (2.1.8), (2.1.9) are violated. 2.1.7 that
H
=

Assuming a perfectly flexible tape, it can be found from Fig.

0.21
,

The stiffness S, has the effect of about 4% reduction in H which will be neglected. The film thickness is then
*

(Fig. 2.1. 1)

r H
0

2/3

=0.25x 10 x0.21x (6.35x 10

-6 2/3

=18 inch

RR 71-18

2-15

AMPEX

2. 1. 2

Determination of the Steady-State Undulations in the Exit Region In the exit region the foil undergoes a series of undulations,

which decay in the direction of the central region.

When the reference

conditions (Eq. 2.12-2.19) apply, the minimum gap is h /h =0.716

min

and the maximum gap is

hmax/h

1.065

The last peak-to-trough angular distance ("half wavelength") is

A8

2 .0

(6(2.1.10)

The effects of deviations from the reference conditions on the undulations are given in Fig's .2.1.8 through 2.1.12. Parameters not ex-

plicitly defined in the figures, satisfy the reference conditions (Eq. 2. 1. 2-2. 1.9). 2.1.3 Remarks on the Steady-State Characteristics of Finite Width Foil Bearings. No complete, finite width theory for foil bearings is presently available. The main qualitative effects are summarized below and some sample
[

results are presented a.

2.4, 2.11].

Foils bent by couples into cylinders display edge-undulations (anticlastic effect), which decay rapidly toward the foil centerline, as shown in Fig. 2.1.13.

RR 71-18

2-16

AMPEX

T-

--

L.

-r
-~

---

-'CD

~7it7-C> ~~70

ol

-4

Ca4 cA *4

W 1

L4.4d

E-4~

-4

II)
VI

II
______________________________________________________________

07

RR 71-18

21

--

AMPEX

0.

1
00
0 0

11

-1-----.--Wo
z

Co

14-4
.

4-3

. 4-3
4~4

0.

.10

________________________________

t16noil______________

RR 71-18

2 -180

AMP12X

Theory
Experiment o
4I-J

2.6
[2.4]

B 1L111
Undulating

14

>-Exit
-a

Region --

Non-Undulating
Exit Region

---

1
-

T pr ao

I
I

C "-

0 0

o,

(tii

0.01

0.1

1.0

10.

Compressibility Parameter

i/C

Fig. 2.1.10 Effect of the Compressibility Parameter 1/C on the Wavelength of the Exit Undulation

RR 71-18

2-19

AMPIEX

1.2 PEAK h max S*

INCOMPRESSIBLE

.00
A 0.8

L
TROUGH., /

hmi n

.C

0.6

UNDULATING EXIT REGION

MONOTONIC
EXI REGIO EXIT REGION

r. (a 0 .4

I .. . ......... .. . 1

0.2

LL1

0.01

0.1 COMPRESSIBILITY PARAMETER

10

Fig. 2.1. 11 Effect of Compressibility on Last Peak and Trough of Exit Undulation [ 2.6)

RR 71-18

2-20

AMPEX

Theory -[2.8) 'Experiment o [2.43 2 /2 pU r" .0 0.1 9-

0.9

0.8

h min h*

0.7

0.6

__..'__h 0.4

min0T

(1.

2.0

3.0

0
(6pU) 1/3 T

4.0

S.(

6.0

7.(0

8.0

9.0

Fig. 2.1.12

Effect of Wrap Angle on the Trough of the Clearance Undulation 2-21

RR 71-18

AMPEX

b.

The effect of circumferencial tension is to flatten the undulation.

c.

The effect of the rapid pressure drop to ambient in the vicinity of the foil edges causes the edge zone to sag, creating a
partial or total seal.

d.

Side leakage decreases the gap width in a circumstantial direction particularly with foils for which b Fig. 1.2.1c.

Fig. 2.1. 14 shows typical tape contours traversed across half the width of the foil at two longitudinal stations. The upper graph shows Comparisons the contour at the bisector of the 300 -wrap angle, whereas the lower graph shows the contour along the locus of minima in the exit zone.

of parts (a), (b), and (c) of the figure give an indication of the effect of foil thickness, for the same nominal clearance. creasing the speed) for the same foil thickness. 2.1.4 Film Thickness of Foil Bearings of Finite Length (Fixed Points of Attachment) In applications of equation (2. 1.1): tU)2/3 h*= H*r ( (2.1.11) The comparison of parts (c) and (d) shows the effect of increasing the nominal clearance (by in-

to transports of tapes and webs, the tension T s is prescribed. In applications to rotating machinery, the tension T is coupled to the elastic stressS

strain equation [2. 12]: Ts - T0

*6

-R - 2ysin- 2 + Oh

(2.1.12)

Et 0 "t is the increment in foil length due to reasons other than stretching

e.g. thermal. y is the component of displacement of the rotor center, measured away from the foil, along the line which under the initial t =0, h* 0) is the foil bisector. conditions (TO , o RR 71-18 2-22

AMPEX

0.975 t

Foil Lateral

Profile

3.07

2b

Moments

--

-Bending .IvMoments

Fig. 2.1.13

Lateral Foil Profile Due to the Effect of Pure Bending

RR 71-18

2-23

AMPEX

0 0.3~-0.s? O0 0.620.771.181.54m.

0.500 -OA75 -0.470 6 0 0.4500 0.425 0.400

0.5 d
w ) C.)

0j0.A

IP .

2.31-

J4J

31-

0.250 WrCL

ci)

.. i

0050

7.

-- - -T
-.40 -0.460
-0.450 0.425 -0.400 U)U w u)Lm co w. o..,
rC,

-4

0.45..A75 0.740.921.391.85-

0.55-

J.

0,
'T

tk

-0

(a

m.2.7T-

-0.350

'C'
0 co

4.2
5

4.62 z0.250 6 z-z

CD LU

t/) :-.4

9.24T
0
-

,
1,
0 d
(D 0 v re0 c(l

0 4d,'
0

"q4/N4 INV

/4

RR 71-18

2-24

AMPEX

"*00.500

0.43 0.530.66-

0.33-

0A0

.s0.475

0.470 -0.460
0.40

0.991.31ax~

-0.425 -0.400 "

.4

CD 0.350 wU . .
D

Q) (n
0.S'1z .0

Fn l

to

a.)
W

3.27

In

0.250
w

-) 6.54
aNV 0).p

~/
0

0.330.40"

"0.500
-0.475 4)

"0

0.530.66

0.30.460

0.470 0.450

U)

0.97-

"5

0.250

' 0G

4.9

3.27--0.2

CU) W

0 f-

( c)

c)

6.54

I .
"lq/Nlq aNV 14q/

RR 71-18

2-25

-AMPEX

The above is a consequence of the fact that foil lengths in continuous transport are very large, whereas the elongation of finite, fixedend foil segments, used in the support of rotors, has an appreciable effect on the tension. The design curves in Figs. 2.1.15 and 2.1.16 give the

clearance h* and the tension 1' for the case when y = 0 (concentric rotor position) and 6-, = 0 (elongation due to stretching only). These curves are presented here only as a guide. The user is

cautioned that the following effects are not included in the curves: 1.
2.

Foil Slack
Fluid inertia

3. 4.

Thermal elongation Slippage of the foil at points of attachment.

desin ofLich

Comparisons with film thicknesses measured in the particular [ 2.13 ] show that Fig. 2. 1.15 underestimates h* by a factor

design of Licht

of 2. When corrections for the above factors are included the predicted values are in reasonable agreement with the experimental results.
2.2 Response to Sudden Disturbances

Disturbances in the contour of the foil, as well as the response of the foil contour to tension step-changes, propagate downstream, from entrance to exit, at approximately r2.14 V p
-

2.16]

(2.2.1)

This speed of propagation Is not appreciably affected by stiffness and

compressibility. Computer simulation of the response is cases in Figs. 2.2.1, 2.2.2. shown for two typical

The film thickness distributions at successive

time intervals are shown by displacing upward by one division the datum for

RR 71-18

2-26

AMPEX

consecutive curves. distribution only.)

(The scale is indicated for the initial film thickness

The speed of propagation of a disturbance produced by an impact of the foil was measured by Licht [ 2.15] who verified the theory [Fig. 2.2.3]. 2.3 Response to Periodic Excitation If a sinusoidal variation of tension, T(t) = T (I + p coswt)
5 0

(2.3.1)

in which p

<< 1, is imposed on the foil, the steady-state response of the

film thickness can be shown to be of the form:

h (E),t)--h

(e)

H*

H(* co ~

cos [wt+ Y (e,wj

(2.3.2)

in which A (p ,w), and y/(e, w) are the space and frequency-dependent amplitude factor and phase [ 2.17]. Typical spatial distribution of the film thickness perturbation is shown, in Fig. 2.3.1. One should note the decay in amplitude and the The spatial and frequency travelling-wave character of the phenomenon.

dependence of A and Y is illustrated In Fig. 2.3.2. The physical interpretation of variation of A and y with increasing frequency of tension pulsation is as follows. The tension variation in the foil can be assumed to be spatially uniform, with the gap responding first in the entrance zone and the front of the change washed downstream at nearly U/2. At the same time, for a compressible fluid, local instantaneous compression is synchronous with changes of tension.
b

RR 71-18

2-27

AMPEX

0 Soo

1
1,

.2
,

.3
I

.4
14

SurfLve Speed (sam~ple) U x 10- (ips) .S b .7 a8 .9 1.0 1.1 1.2 !.3

1 I I

0.45 x 10 ) 0 ,o 1.0~
Ur

P,

1 0

'pp.i

- .499 x 10#
(t 0.001"
0 boop

' '

r)

~
acap/P

119
-.

.7

~-4--.6

PT _

~cc

a e

Sped (sa Su~i~e

0le)6 x 10

(ips

0a
r.0
(T,/P Wi.0nbIn 4.0 1 .24 .3 .7 . . . 1 1

-. 2
Win

UU

Fig. 2.1 .15a

Effect of Foitil Ticnsso on Steady State Clearance [2.12]f

RR 711

2-2800.8

AMPEX
P Surface Speed (sample) U x 10 4(lps) .0 1.1 1.2 1.3 .7 .8 .9 .5 .6 , 2

0 4oo

.4 .1 .2 .3 +l PT-o o o

200

(r

.0

----

--

1.-04

P
U

400

p.01

Fig. 2.1i. 15c Effect of Joumnal Radius on Steady State Clearance [2.12 ]
Surface Speed (sample) U x 0 lps) 1.1 -1.2 C~s 1.3 3ocw~ 77I "l.0 .2) . 1.2 -

-=
-

500 0

.1

.2

.3

.4

.5 P,

.6

.7
t o_

.8 l

.9

1.0

IUu
r
(00 0 -

P =0.485x0 0.49 x 109

Ph

=
-

400 o _. = ...... P 0 .1 .2 . 0 .4 5 .6 .7 .8 .9

t =0.001") 1.0 1.1

Fig. 2.1.15d Effect of Wrap Angle on Steady State Clearance 2.12)

RR 71-18

2-29

AMPEX

.1 , OAK

S+rt.,, Speed (smaP10. U x 10 1.0 .8 .9. .$ ,6 .7 .. .4 ,! P '

ps) 1.1

r0
-

1.2 i

1.3

Et,
Ir

-- a

O -,00

"

t ft"Q"

aCas

fTo
i \

(.

T
: o.(, x 10

fa 7
!

')
.

{T ., Lb'ib) ', .4 .5 , .7 .8 .9 1.0 .1.1 1.2

Sped Parameter Pu

Fig. 2.1 .16a

0 .1

Effect of Initit-I Tension on SteadyState Tension t2'.12,].

.2 .3 .4 (lps) Surface Speed (sample) U x 10 .9 1.0 1.1 1.2 .5 .6 .7 8
- 4

p 1.3 1 u
'

P -_P

I
-. 4

a3
-10

P(t

00 1 0.97 x10

.9

x .

..

P1 Referenc

.
I

9
8/a

Pr

T 0 P' P

a/p a

.2

bl

0
, \PI , O.A

7
x 101

- - - ,. oo
0

,.,
3

I..

.1

ft .2

0.00511)

60~

,3

.4

.5

.6

.7

.8

.9

'1,0

1.1

1.2

Sneed Parameter P.

Fig:. 2. 16b Effect of Foil Thickness on Steady State Tension [ 2.12]

RR 71-1.8

2-30

AMPEX-

.5 0 a

.1

.2 a

.3 aV

.4, aI

Surface Speed (sample)'U x 10 lips) .5 .6,.7 *J .8 .9 1.0 i.1 a a U 6 r = r === PT

1.2 '

1.3

0.499,x (t=0.001-) 10,Iu

PTo- O.C'61x 10' 5 0_ TC- 2.0 b/in; ! *

r ~P"= ~ 0

E- --L
P/
P .T . . a

14
3

.4

12
-

1-.

-0-i
i
r,
(r r

-9

C .

22
PP , 0.4 I x.8

t
2

Pr =0.485 .x

.{

.3

.4

.7

.8

.9

1.0

1.1

1.2

Speed Parameter P.

Fig. 2. 1.16c Effect of Journa!: Radius on Steady State Tension [ 2.12]

Surface Speed (sample) U x 10 (ips) .5 .6 .7 .8, .9 1.0 1.1 1.2 , a, , ,a u .- pr U P, r o Et a r l J a a i P T T" a - i2 ,-ta -!0
- 4

.Sa 0 r1

.2 ,

.3 ,

.4 ,

1.3

- !

P - 0.485x 10 .4 ,(T1 xu (r0 z 1.08V * PTo 0 .66 x 10 0

o

1I

2.0 Wbin,) 0.499 x 10

P'

2-

0.0

A-

_ _ _-_
4 3 .2 .I . 3 4 . . 6 .7e .9 10 . . -j

2
78

.1

.2

.3

7 .5 . Speed Parameter

.9

1.0a 1.1

1.2

Fig. 2. 1.16d Effect of WrapAngle on Steady State Tension L2.1 2)

RR 71-18

2-31

AMPEX

POINT OF TANGENCY

j
I

ut -TI - !,,,ro =T ,Pu

'-2r

'TI

-,'3

4.0-

00i
2.2.m a r

2.5.

foil bearing to a saw tooth wavelet of width 2a = 1014 11 .8 and height A - 0.4 (Rapid initial smoothingI of the disturbance occurs initially and is followed by a downstream movement of the disturbance at U/2) [2.14] . Note: In this figure t represents time.
RR 71-18

2-32

AMPEX

'

_u
r=Ut

. .,
GPU
T
-

POINT OF TANGENCY 2 2ro

!, h

i ;H :,i,

=-!
T o

&t

,,
I

4-5
475 4

450 5.0, 425 40,0 H

-------. -

, '-.-'----.,
--F

3Z5350 325 .300.''

'
,

4.0 -

,,

I
,,,

U
--

2,

-.

25.0

30-

22.5 I
20.0

''

. 12.5 2.5
-__5.0

20

~
.

'0/(.0
. 5.0
-

.. ... .

..

-,

.. .

.. .
-

..

... .

. .. .
.

I'
.

-M

2.5 ...

2 0-2-4 -6 -8 -o

-10-

-6-4 -2

Fig. 2.2.2 Computer simulation of the dynamic response of a foil bearing to a step reduction in tension. (Greater film thickness is established by a front propagating from the inlet towards the exit at U/2). [2.14]. Note: In this figure t represents time.

RR 71-18

2-33

-U/2 873 In. /sec: VP 850 in. /$cc pC

CC

AMPEX
U 1746 in. /sec- Long Pulse

II

0-

7%

Icm

=0.5

2n mcs10m

U/2

2162 in. /sec;

VP

00 in. /sec

I1cm 218 Fig. 2.

0.21 m of ofaDsubacPln of

3/c In.cit Coprio 2020re Prpgto

dihteTertclyPe

Foi

ditdVlefU221

RR7-823

AMPEX

4-4JQ

LLO

AA

.-

4-4

O*

-1 0

000

000

41-

4
4)

~0
z;

woco

4'

44

z
RR 71-8

)
2-3

.4-'

AMPEX Q)

IXN

21
44

0o

0. 3a

14

U 0

0 L.1
-40

L. -l

0
- 4

a, U)

.2
4.i

RR 71-18

2-36U

AMPEX

Thus, when the frequency is low and the wavelength large, the displacement of the foil in the region of wrap is nearly uniform (P- = 0.1). With increasing frequency, the number of travelling undulations in the region of wrap increases. With regard to the phase y, low frequencies correspond As the

to quasi-static changes of gap, associated with negligible Iag;

frequency increases, the delay for the effect of the the tension-pulse to reach a point downstream of the entrance increases. For an incompressible

fluid, the increase is linear, but compressibility introduces a periodic component in the y curve. The foil approaches its straight-line asymptotes from above, away from the entrance zone, and from below, away from the exit zone. An increase in tension causes the foil to move toward the asymptotes, essentially parallel to itself. Thus, the foil excursion leads the tension

by 1800 in the approach to the entrance, while the foil perturbation downstream of the exit lags by 1800 the motion at the end of the angle of wrap.

RR 71-18

2-37

AMPEX

REFERENCES 2.1 H. K. Baumeister, "Nominal Clearance of Foil Bearings," IBM . of Res. and Dev., Vol. 7, No. 2, April 1963, 153-154. 2.2 A. Eshel and H. G. Elrod, Jr., "The Theory of the Infinitely Wide, Perfectly Flexible, Self-Acting Foil Bearing," J. of Basic Eng., Trans. ASME, Vol. 87, Ser. D, No. 4, Dec. 1965, 831-836. 2.3 J.T.S. Ma, "Investigation of Self-Acting Foil Bearings, " Basic Eng., Trans. ASME, Vol. 87, No. 4, Dec. 1965, 837-846. 2.4 L. Licht, "An Experimental Study of Elasto-Hydrodynamic Lubrication of Foil Bearings," T. of Lub. Tech. Trans. ASME, Vol. 90, Ser. F, No. 1, 1968, 199-220. 2.5 A. Eshel and H. G. Elrod, Jr., "Stiffness Effects on the Infinitely Wide Foil Bearing," T. of Lub. Tech., Trans. ASME, Vol. 89, Ser. F, No. 1, Jan. 1967, pp 92-97. 2.6 A. Eshel, "Compressibility Effects on the Infinitely Wide, of

Perfectly Flexible Foil Bearings," Trans. ASME, Vol. 90, Ser. F, No. 1, Jan. 1968, pp 221-225. 2.7 W. E. Langlois, "The Lightly Loaded Foil Bearing at Zero Angle of Wrap," IBM J. of Res. and Dev., Vol. 7, No. 2, April 1963, 112-116. 2.8 E. J. Barlow, "Self-Acting Foil Bearings of Infinite Width," J. of Lub. Tech., Trans, ASME, Vol. 85, Ser. F, No. 3, July 1967, 341-345. 2.9 A. Eshel "On Fluid Inertia Effects in Infinitely Wide Foil Bearings," J. of Lub. Tech. Trans ASME, Vol. 92, Ser. F, No. 3, July 1970.

RR 71-18

2-38

AMPEX

2.10

A. Eshel "On Controlling the Film Thickness in Self Acting Foil Bearings,"J. of Lub. Tech. Trans. No. 2, April 1970. ASME Vol. 92, Ser. F

2.11

A. Eshel & H.G. Elrod, Jr. "Finite Width Effects on the SelfActing Foil Bearing," Report No. 6, Lubrication Research Laboratory, Columbia University, New York, N. Y., 1966.

2.12

A. Eshel, "Dynamic Analysis of a Three Foil Rotor System in Zero Gravity Environment," J. of Lub. Tech., Trans. ASME, Vol. 92, Series F, No. 4, 1970, pp. 617-629.

2.13

L. Licht, "An Experimental Study of High Speed Rotors Supported by Air Lubricated Foil Bearings," I. of Lub. Tech., Trans. ASME, Vol. 91, Ser. F, No. 3, 1969, pp. 477-505.

2.14

A. Eshel, "The Propagation of Disturbances in the Infinitely Wide Foil Bearing," J. of Lub. Tech., ASME Trans., Ser. F, Vol. 91, No. 1, Jan. 1969, 120-125.

2.15

A. Eshel & M. Wildmann "Dynamic Behavior of a Foil in the pr-sence of a Lubricating Film," Trans. ASME, Vol. 35, Ser. E, No. 2, June 1968, 242-247.

2.16

L.*Licht, "On the Velocity of Progration of a Distrubance Along a Foi],"j. of App. Mech., Trans. ASME, Vol. 36, June 1969.

2.17

T. Barnum and H. G. Elrod, Jr., "A Theoretical and Experimental Study of the Dynamic Behavior of Foil Bearings," Columbia University, Lubrication Laboratory Report 14, 1968.

RR 71-18

2-39

.5

AMPEX

RR711

24

AMPEX

NOMENCLATURE (Chapter 3)

Orifice cross-sectional area

Effective orifice cross-sectional area (a, Cross sectional area of capillary

a' a

c a

do0

b cd H* 1 H 2 ,c , L gh*
Ps Pg r
0

Foil half width Discharge coefficient Film thickness upstream of pressurization groove Film thickness downstream of pressurization groove Length of capillary Circumferential length of pressurization groove Dimensionless groove length = Y. ( 1
Supply pressure Groove pressure Radius of foil bearing journal (guide) Q1 Inflow through entrance zone Outflow from bearing Inflow through groove restrictor Foil tension per unit width Algebraic sum of journal and foil surface velocities Speed of propagation of disturbance

Q2 Q T U V

RR 71-18

3-1

Preceding page blank

AMPEX

Ah* AP

Change In nominal film-thickness due to pressurization Pressure parameter, &P = h1 *

Foil bearing number
=6U

ro(pg-Pa)-i

/(ro2/3

RR 71-18

3-2

AMPEX

3.0 EXTERNALLY PRESSURIZED FOIL BEARINGS

3.1 3. 1.1

Steady State Characteristics Interaction of Self-Acting and Pressurization Effects In a variety of applications, it may be desirable to supplement

the film thickness generated by self acting effects, by means of external pressurization. Conversely, it may be neceszary to withdraw fluid from the lubricating film. A representative configuration is illustrated in Fig. 3. 1. 1 show1.

ing a lateral groove in the region of wrap, extending over the major portion of foil width. uniformity. This type of foil bearing is characterized by two regions of Here, in addition to the entrance and exit region at the ex-

tremities of the angle of wrap, secondary exit and entrance regions form in the vicinity of the groove. The flow through the region h* 1 is augmented or diminished by the flow through the groove restrictor, so that the clearance h* 2 in the downstream region of uniformity can be greater, or less than the clearance h* i" The design of this bearing can be accomplished with the aid of Fig. 3.1.2. The gap width in the upstream region of uniformity can be Assume that it is required to increase the gap h + Ah*. We enter the design and

evaluated from Eq. 2. 1.1.

width in the downstream region to h

chart in Fig. 3.1. 2 along the ordinate of hthe required value of Ah*/h* 2 1 *+A*Weetrtedsg locate the intersection with a (Lg = 't
a selected groove width
. .

I/3/h* )-curve, corresponding to

Proceeding now horizontally, we determine

the groove pressure pg from the value of the pressure parameter

RR 71-18

3-3

0.0

o
o
0

o
0

E-

+
C

o00

Ol

0
0 C4

0
04

(1)

00

t l

4-JI

r,
4

-4
R7183

0
E-'

U,

.-

u~C.

RR 71-18

3-4

AMPEX

. 16

0.14
0.12 < 0.10 L0 . 0 8

Oe' -"

"

L =0

0.0

I/f

2 3

0- 0.04
S0.02
0

-002 -004-006

-008
-0.10 -1 0 1 Ratio, 2 3 6h /h 4 5 6

Fig. 3.1.2 Computer Simulation of the Clearance Response of a Foil Bearing to a Moving Supply Source

RR 71-18

3-5

AMPEX

(3.1.1)
The second part of the design procedure involves the determination of the supply pressure and type of restrictor compatible with the bearing flow rate. The outflow from the bearing Q2 is the sum Q + Qg of inflows

through the entrance zone and through the groove restrictor (assuming side leakage to be negligible). Symbolically:

Q2

Q1 + Qg

(3.1.2)

or

2b(Uh*/2) = 2b(Uh* /2) + Qg 2 1 g so that the flow entering from the groove is: Qg= bUAh* g in which U and A and Ah* = h* width. If the groove restrictor is an orifice: Qg= cdao a/2(p s Pg)/P 2 - h* 1

(3.1.3

(3.1.4) are known and b is the foil half-

(3.1.5)

With Q known and p previously determined, the supply g g pressure ps can be selected for an effective orifice flow area a = cd a If the groove restrictor is a capillary, Qg= acd 2(p - pg)/32 g c s g t, (3.1.7) (3.1.6)

RR 71-18

3-6

AMPEX

Again, with 0

and p known, the supply preu sure ps andi the

dimensions of the capillary, a c and 3.2

.C ,

can be selected.

Response to Step Disturbances When a sudden tension increase (or reduction) is imposed on a

self acting foil bearing, a corresponding reduction (or increase) of film thickness occurs immediately in the inlet region. gates downstream approximately at Vp = U/2. This change, then propaIf a pressurization groove,

separating two uniformity zones is present, wave-fronts, travelling downstream at nearly U/2, will be initiated in the entrance zone and at the trailing edge of the groove. When the inlet disturbance reaches the groove it becomes a "new" disturbance for the inlet of the second zone of uniformity and is washed downstream again at a speed Vpz U/2 Fig. 3.2.1 p 3.3 Response to Moving Sources In the application of foil bearings to the supoort of high speed rotors there arises sometime the need for external pressurization. If 3.2)

pressurization is through the interior of a rotor, the system corresponds to a series of moving sources. The response of the clearance to a rotating [ 3.2 ] pressurization groove is shown qualitatively in Fig. 3. 2.2 To begin with, the clearance profile is that of the self acting bearing. atmospheric, must rise to pa + T/r in the region of wrap.
a 0

As the recess

approaches the inlet zone, the foil dips since the recess pressure, initially When the recess is filled with gas to the bearing pressure-level, then, as it proceeds downstream, the fluid supplied causes the foil to bulge. The front of the bulge travels downstream at the speed of the source U, whereas the rear of the bulge travels downstream at approximately U/2. leaves the zone of wrap. Thus, the length of the bulge increases at a rate nearly equal to U/2 also, until the source eventually

RR 71-18

3-7

AMPEX

POINT 01: TANGENCY

7.0

"P-

6.0

40.0

3.0

2.0

0.0

Fig. 3.2. 1 Computer Simulation of Dynamic Response of an Externally Pressurized Foil Bearing to a Step Reduction in Tension RR 71-18 3-8

AMPEX

PUWTr

OF TANGENCY
-VJ1/3

7
IT=
40.0

i
f

'36.0.

32.0
9-2' D
--

'1,i160~

102D

42 0 -2-4-6 -8-10

-10 -8 .6 4 -2 02 4

Fig. 3. 2.2 Computer Simulation of Dynamic Response of a Foil Bearing to a Moving External Pressurization Source (marked with broken lines) RR 71-18 3-9

AMPEX

REFERENCES

3.1

E.J. Barlow "Interaction Between Self Acting and Externally Pressurized Effects in a Foil Bearing," Ampex Corporation Research Report RR 65-12.

3.2

A. Eshel and E. J. Barlow "Static and Dynamic Analysis of Externally Pressurized Foil Bearing." In preparation 1971.

RR 71-18

3-10

AMPEX

NOMENCLATURE (Section 4) E fe f F,F x y G Young's Modulus Frequency of exditation Undamped natural frequency f =mResultant force components on rotor in x, y directions respectively
x,y

Amplitudes of transverse and in-line excitation (in g's) Film thickness of foil bearing in uniformity zone Dimensionless film thickness in uniform.ity zone H* = h*/(r 0 (See Chapter 2) Polar moment of inertia of rotor Transverse moment of inertia of motor Subscript denoting kth foil sector. width. Initial foil length (at To) Rotor mass per unit width of foil Rotational frequency
Speed parameter P
=U

h* H* Ip I k to m N
P

Bearing stiffness per unit

r0

Pr

Radius parameter

pr =
ro0

Et

Length parameter

RR

a8

RR 71-18

4-1

-AMPEX

PT
F~I

Tension parameter

PT
i

T P/P a a
C

Dimensionless damping per unit width

= c/1.

Ph

Dimensionless gap

Ph

'J/P /paa

Dimensionless frequency

Pf =

I fe/p

P P p a r
K

Dimensionless stiffness per unit width Frequency ratio Ambient pressure, absolute Journal radius Tension per unit width in k th foil sector Preload tension per unit width Foil thickness P = ro/U

k/pa

Tk T t U W x,y Y k

Tournal surface speed Rotor Weight Components of displacement of rotor (Fig. 4.2. 1) Component of displacement of rotor away from and along the bisector of the k th foil sector. (Fig. 4.2.1) Angular position of k th foil sector relative to the first foil sector (Fig. 4.2. 1). Elongation of k th foil due to reasons other than stretching (e.g. thermal). 4-2

Yk 6 k

RR 71-18

AMPEX

p. Pa

Viscosity of lubricant Density of lubricant at ambient pressure

Wrap angle

RR 71-18

4-3

AMPEX

RR 71-18

4-4

AMPEX

4.0 FOIL ROTOR SUPPORT

4.1

Advantages of Foil Bearings The application of foil bearing technology to the support of

small, high-speed rotors has been shown to be practical and to offer [p4.1-4.3)] certain advantages over other types of gas bearings. . These advantages are: 1. 2. 3. 4. No whirl instability. No sensitivity to fractional frequency excitation. Self alignment. Simplicity of construction and low requirements of manufacturing accuracy. 5. Accommodation to thermal gradients. 6. Excellent wipe-wear characteristics and tolerance of foreign particles. 4.2 Steady State Characteristics The procedure of determining the load capacity, the tensions and the gaps for a given displacement (x,y) of the rotor is as follows. 1. Flndyk (k= 1,2,3; Fig. 4.2.1) Yk
=

x sin k

+ y c os

Yk

(4.2.1)

RR 71-18

4-5

Preceding page blank

AMPEX

2.

Solve for Tk and hk* simultaneously by iteration of the following equations.

T k - T 0 -61t k - 2yk sin0- +
2

(4.2.?)

Et

-,
0

U6(4.2.3) h*= H*r k oTk-/ k 8 k

(see Sec. 2.1.1)

(4.2.3)

elongation of the k th foil due to reasons other than stretching e.g. thermal.

3.

The forces on the rotor are -(T F=V3-si 3 - T2 ) VsinF = 2 sin-

(4.2.2) (4.2.3)

4.3

Stiffness and Damping Coefficients

Theoretical curves of the stiffness and damping coefficients for a self acting foil rotor support of simplified geometry, shown in Fig. 4.2.1 are given in this section. These curves apply to the case when no radial load is applied to the journal. Though they do not constitute a complete map, they include a range of parameters which is of practical interest. The following simplifications pertain, to curves presented herewith: a) The clamping method of foil ends is idealized, as shown in Fig. (4.2.1).

*The length of the foil is approximated in this formula as composed of two straight sections and a circular arc of radius r + h*.
o

RR 71-18

4-6

-iAMPEX

3 rd Foil Sector

Point 0 is the No-Load, No-Rotation Equilibrium Position

,:

3 0Y

yy

1st Foil Sector

Y3 Q1

R2, 2nd Foil x \IkSector

Fig. 4.2.1

Schematic Diagram of Foil Bearing Configuration

RR 71-18

4-7

AMPEX

b) Fluid inertia effects are neglected. c) The relaxation of foil tension due to temperature increase with speed is not taken into account. d) The slack remaining in the foil due to its finite stiffness, is neglected. The effects of foil thickness t, Journal radius r , wrap angle O preload tension T0 , rotational speed N, and excitation frequency f upon the stiffness coefficient k and the damping coefficient c are shown in Figs.
(4.3.1) - (4.3.4). The stiffness coefficient may be estimated on a quasi static basis by means of the formula:
2 6 Sin 2

- Et

2 -

0Eau 0r o r
0

r2Et +_+9(I

aT

e (6LU)/ T T/Et

2 3.3

H* - ___

(4.3.1)

aT

Results based on this formula are indicated along the Pk-axis (Pf 0) of Figs. 4.3.1 and 4.3.2. The damping coefficient cannot be es-

timated on a quasi static basis. A comparison of theoretical and experimental results is shown in Fig. 4.3.5. The curve marked theoretical I is based on stiffness calcutaken as zero. The next three
(b), and

6_t. H lated from Eq. 4.3. 1 with _ 6T and a3T

curves are obtained by sequentially removing the idealization (a), (c) above. 4.4

Design Considerations Related to the Bearing Frequency-Response It is noteworthy that for a foil bearing the damping coefficient

vanlshes whenever the ratio of excitation frequency fe and rotational

frequency N is an integral multiple of (Fig. 4.3.3, 4.3.4).

?/e,

where e is the wrap angle

The bearing stiffness, however, is not sensitive to The following points out

either the excitation frequency, or to the speed. at which the damping vanishes. RR 71-18 4-8

some design implications due to the existance of these "critical" frequencies,

0.30 "~1

-----

r 0.242 10 |o - 0.S

....

I
C/ 997 x 09

AMPEX

- 4.0 .

0.28
0.26, oP
/

111

3."_

1
3.4
3.2 r ffect ofroil l'ickness W~ect o!Roto Radxws3. ;icm ., V iaial ie~m 3 .o
-

..
0.24 /
0.22
__ _ .

00

~~Lffect
_ _ _ _ _ _-

Aproxinate A.,ympl':: ; 1.e (E. 4 3.21'

Fe fe m ne C a b e c 2.t -"

. 1e

oae

I
f

l,
uP T0

- 0.6 ,, (T. ,i , O ' 2.o

1.5 l ( RS - 0.499 X 0~(t- .O01 "

!
2.-

,P

P PT.

(o4.lb m
Cd,

1.32 x I0-

- 0.48Sx10

(t

1k.0.,

2.2

.0

1.6

o.,o"

SPt 0. 1,49 10 x

--9 -

0.0

-I--, _ 1.2 1.0

\
'
0.04

_________ ___
aa r
o

.
P ., t
ta

0 V' X I0 O .-j;
(o

Pft

7 P

0. 02

. --- ..

II . --. . .

~~~
24

~~rlr*."
rpq sncv P, x 10

Pa/JIID,

12

Fig. 4.3.1

Stiffness Coefficient of a Foil Rotor Support

with Zero Radial Load as a Function of Excitation Frequency, Foil Thickness, Rotor Radius and Initial Tension RR 71-18 4-9

AMPEX

_____

4.0

e~~~~.900___1

________

0.24

3.4

0.22

r.

3_2

- 0.20

;.Effect
________Approximate

of Wrap Angle LEffect of Rotational Speed

A____ptotc Value
6

PT
-

T
3.0

P___k____

. 8

_a2.6

.0.227 v

x 10

ILL U 01 .
2.2

0.16

(200 RPS) .

2.4

3:

i u.

0-

0.12

_____

____

___

____1.8

0.681 x 10x (600 PE'S)

R31eOnco Case

(,

U
c

to

-' 0.66 X10

-0.499x10
_._I____ _...._ _

~
.

2.Ojb/in)

1.4
1.2

(t-0.001*j
5

Pu/Pr

,
-

0.452 x 10 6 (400 RPS

0.-85x10 (ro 1.0")

I
0.06 [ 0.04 36o_.__ __

(r-1.0o

1.0
8

_-

_ --. __

__

0.02 7! ....

.4

12

24 20 I, Dimensionless Frequency Pfx 108

28

32

36

40

44

Fig. 4.3.2 Stiffness Coefficient of a Foil Rotor Support with Zero Radial Load as a Function of Excitation

I,

Frequency, Wrap Angle and Rotational Frequency RR 71-18 4-10

AMPEX

.o

.00

t- P 2.2

I
.08c

<

TN

4
022

;j

rr t
2
.F

2
r.a - 2 0 1/

, rr LO 2 0 .

'PT

Effe ctnOf r ol T hik ne s

.1

-_____________________"_______________
CC

I Refecnc
pt . 6 - I oa

oa Zer Rdil asa

T 0- 0 .0g 05i Referseonrcedusan g Frqeny

untinff eci.tati".Aon
to Eff R f S u t h

n oe f ci _Ain

0R

1!

Zero~~AS Raia Load as a Fucio.f0xitto Foi Thcnes Roto Radisan Frequency,48 Iniia Teso843 6

44

RR71104l

AMPEX

I_
_ _

34

1
t
.
-. -

_3.2

.11
/

_______

13.0
2.8
d

..
.!

Effect of W ap An;Ie Effect of Rotaor 1 Sp

I
.09

-,.

2.6

1.
"P/P/.

_2.4 _

0.66 X10
=0.499

(T - 2.0lb/1-)

x10 (t -0.0011

.08 -'

- 0.452 x i0o6 (400 RPS)

E~

0.02

10

"

60 " 0.48

6 '...

18 2.0-C

o(200 -0 .2 2 7 RPS) 10 ;-7" pa~

1-.

.06

"

-1.6

.04
.03

Cis.4

I
-

I
1
6

I
P -.

,r2
-

.03--

0.681 X 10(600 RPS)

0.

.-

.-~

---. H-.--L
-0.4

.01

Loda-Fntono

xitto

reuny

Wrp

n-

n
- .

-0.2

Fig 4 3.

12

te0(

U$,

Fig..3.4Dam ping Coefficient of a Foil Rotor Support with Zero Radial Load as a Function of [Excitation Frequency, Wrap Angle and Rotational Frequency 4.3] RR 71-18 4-12

AMPEX

300

300.. 3Theoretical

. I

250

_____.

> 200

2150 m
.4-'Theoretical

'

II

__

Material

Steel

50

Thickness (Mils)

1.0

Preload, T (lb/in)
0 0 100 200 300 400 500 Rotor Wt. W(lb) 600 700 800

2.0
2.414 900 1000

Speed - N (RPS]

Fig. 4.3.5
]f,

"Natural Frequencies" of Foil-Bearing Supported

Rotor; - Comparison of Theory and Experiment

[4.4]
RR 71-18 4-13

AMPEX

, and the excitation F N m 21T frequency T occur at a rotational frequency N such that

If both the resonant frequency f -

f
n

T Ni

i=

1, 2 ...

(4.4.1)

there will be no damping to limit the resonant amplitude of the system. When the excitation is synchronous, the foregoing condition arises only when purposes. For any given operating speed N, this coincidence is avoided irN rotor mass is such that f is considerably less than

e=

ir , which is a wrap angle impractical for design

if the

During acceleration and deceleration, however, points at which

f
n

N\

~,

2 ...

If the bearing is excited at f e e speeds a dangerous situation may occur. are always encountered. Example
Assume e N k m = 600 = ff /3 = 400 rps = = 10,000 lb/in/in 3 Ib/in rad.

f n at these

The resonant frequency in the planar mode is 1 27T ~ m 1 21T 100 3/386
18 p

n etc.

The excitation frequencies with zero damping are fe = 1200 cps, 2400 cps

RR 71-18

4-14

AMPEX

At the operating speed, therefore, no danger exists.

f =f
e n

On the

other hand during acceleration or deceleration, an external excitation at = 180 cps will be dangerous, whenever the frequency of rotation is
fn

N=--= where i=1, 2 ... 4.5

I Fo -=

60,30 ...

rps

Example of a Turbo-Alternator Bearing Design

Fig. 4.5.1 shows the essential components of a 21-lb simulator supported by foil bearings 4. Starting and stopping is aided by external pressurization, through the interior of the rotor and rows of orifices. This particular rotor was operated stably at speeds in excess of 50,000 rpm in both the vertical and horizontal attitudes.
Fig. 4.5.2 shows a scan of response of the rotor to rotating im-

balance. A scan of reponse to unidirectional excitation is shown in Fig. 4.5.3. The motion of the rotor at various levels of excitation is illustrated in Fig. 4.5.4.

RR 71-18

4-15

AMPEX

LEOEND ROTOR F L EARIN.G

3 FX BEARING
4 5
6 7

9(roaersed)
,

S PPORT FO!L-GUiE POST -FOIL" LOCK THRUST BEARING "THRUSl"-BEARING.,

AIR TURBINE

,
,

II

10, ALIGNMENT BAI 12 ROTOR COUNTERBALANCE 13 1BASE

14 CAPACITANCE PROBEL 15 OPTICAL TACHOMETER ia ALIGNMENT SUPPORT ALIGNMENT PIN , 14

9 NOZZLE RING

4/

105

Fig. 4.5.1 Schematic Diagram of TurboAlternator Simulator [ 4.2]

RR 71-18

4-16

-~AMP*EX

1 2.5

AI

BA

Lzu

P-

K-

W=2O.s lb

IP 0.0595 in/I b/sec2

2 I4I~8S in/lb/sec,

iso

300 450 600 750

SPEED-N [RPS] P=PRESSURIZED, S= SELF-ACTING

Fig. 4.5.2. Scan of Response to Remanent Imbalance in Vertical Attitude (Decreasing Speed) [ 4.2)1 RR 71-18 4-17

AMPEx

ca.

co~t
-I,-

ci4-

00
00t
-

04

000

00

o "I1*
Ch w

1~K
", AE
co~,

2~~i

'~

CO.)-

1
9-n

<
i H-J

Urn
Malmo

ONCO rIxI

-4
l, w 0

4-1

AMPEX

C', .00

w)Q)

C~jC

4-3

UE

Wi

00

co

>

EBE:111'11
RR 711 4-190

AMPEX

REFERENCES

4.1

L. Licht,"An Experimental Study of High-Speed Rotors Supported on Air-Lubricated Foil Bearings," Part I: "Rotation in Pressurized and Self-Acting Foil Bearings," Part II: "Response to Impact and to Periodic Excitation," Tournal of Lubrication Technology, Trans. ASME, Vol. 91, Ser. F, No. 3, July 1969.

4.2

L. Light,"The Dynamic Characteristics of a Turborotor Simulator Supported on Gas Lubricated Foil Bearings," Part 1: Response to Rotating Imbalance and Unidirectional Exertation. Operation with Heating and Thermal Gradients.
pp. 617-629.

Part 2:
%

Journal of

Lubrication Technology, Trans. ASME, Vol. 92, Ser. F, No. 4,

4.3

A. Eshel, "Dynamic Analysis of a Three Foil Rotor Support System in Zero Gravity Environment," journal of Lubrication Technology, Trans. ASME, Vol. 92, Ser. F, No. 4, pp. 630 660.

4.4

L. Licht and A. Eshel, "Study Fabrication and Testing of a

Foil Bearing Rotor Support System," NASA CR 1157, Nov. 1968.

RR 71-18

4-20