tilting_pad_thrust_bearings_performance_directed_lubrication

8/7/2019 Tilting_pad_thrust_bearings_performance_directed_lubrication

1/10

Paper 13

TILTING PAD THRUST BEARINGS:

FACTORS AFFECTING PERFORMANCE AND

IMPROVEMENTS WITH DIRECTED LUBRICATION

M. K . Bielec* A. J. Leopardj-

The effect on flooded tilting pad thrust bearing performance of a number of external variables is examined.At sliding speeds between 10 and 100 m/s, and for specific pressure between 15 bar and 55 bar, measure-ments were made of oil film thickness, bearing temperature, and power loss for various oil inlet systems,oil quantities, housing pressures, a nd degrees of misalignment. Power consu mption in high-speed thru stbearings can be safely reduced by the use of directed lubrication with a drained casing, bearing temperature

being reduced and oil film thickness increased.

I N T R O D U C T I O N

TILTINGPA D THRUST B E A R I N G S operating at high slidingspeeds (in excess of 60 m/s) may fail owing to high padsurface temperatures. Th e object of the work describedin this paper was primarily to investigate the effectivenessof current methods used for the design of high-speed

assemblies, and to evaluate improvements. The paper alsostudies the effects of speed, load, oil quantity, oil pressure,misalignment, and oil flow arrangement on a flood lubri-cated double thrust bearing. The final section considersthe problem of power loss in high-speed tilt ing pad thrustassemblies and shows how significant savings can be madeby using a different system of lubrication (directedlubrication).

E X P E R I M E N TA L E Q U I P M E N T

Th e Orion test rig shown in Fig. 13.1 was used for themajority of the work described. A 124-mm 0.d. doubletilting pad thrust bearing assembly was used, in which eachthrust ring incorporated eight white-metal faced, steel

backed, centre pivoted thrust pads, 28 mm wide. T oenable accurate friction torque measurements to be made,the double thrust bearing was supported in a test headmounted on hydrostatic journal bearings with the thrustload applied by means of hydrostatic pistons. The 55-mm

Th e M S . of this paper w as received a t the Institution on 30 th Nove m-ber I969 and accepted for publication on 23 rd Janu ary 1970. 33

,-* Research Engineer, R . an d D. Dept , The Glac ier Meta l C o. L td ,Alperton, Wem bley, Middlese x.

, Manager, Til t ing Pad Bearing Section, T he Glacier Me tal C o.Ltd , Alperton, Wembley, Middlesex.+

test shaft was driven at speeds up to 15 000 rev/min bya 56-kW variable-speed d.c. motor. Lubricating oil wassupplied to the test head from an external circuit whichincorporated heate'rs, coolers, pressure and temperaturecontrols, etc., to maintain the lubricant within the requiredparameters for each test. Shaft speed was measured with

an electronic counter, and friction torque by means of aspring balance.

For the early tests the bearing performance was esti-mated mainly by means of copper-constantan thermo-couples embedded within about 0.05 mm of the bearingsurface. Later tests used film thickness probes embeddedin selected thrust pads. These probes were of the capaci-tance type with the signal being read on an inductiveratioing instrument (I)+. U p to eight thermocouples wereused in some pads, and a typical set of instrumented padsis shown in Fig. 13.2.

For many of the initial tests on t he directed lubricationsystem of oil supply described later in this paper, arelatively simple test rig (Scylla) was used. This rig per-

mitted larger thrust pads (68 mm) to be tested on a 750-mm 0.d. t h m t Cbllar at speeds up to 3000 revlmin, whichenabled a mean sliding speed of 100 m/s to be achievedcompared with the maximum of 76 m/s that was possibleon the Orion rig. T o minimize power requirements andto simplify loading arrangements the Scylla rig incorpor-ated two pairs of thrust pads, each pair being hydraulicallyloaded against opposite sides of the thrust collar so that

$ References are given in Appendix 13.1.


2/10

M. K. BIELEC AND A. J. LEOPARD

Hydrostatic thrustb e a r ~ n gfo r loadina ,

Test double- /

t h ru s t b e a r m \ n Hydrosta t ic

Fig. 13.1. Orion tes t rig

Fig. 13.2. Test thrust bearing wi th instrumentation positions

there was no resultant axial thrust. Instrumentation con-sisted of thru st pad temperature thermocouples, oil flowquantity, pressure, temperature, shaft speed, and appliedthrust. It was found that results obtained on this simplerig could be subsequently confirmed on the Orion rig.

All the work described in this paper used a mineral oilwith a viscosity of 41 cS at 50C, 8.4 cS at 100C. Themajority of the tests were carried out with an oil inlettemperature of 50C.

G E N E R A L C H A R A C T E R I S T I C S O FT I LT I NG P A D T H R U S T B E A R IN G S

The work described in this section was aimed at confirm-ing, under controlled test conditions, the effects of differ-ent operating parameters on the performance of thebearing.

Effect of speed and load

Fig. 1 3 . 3 ~shows the effect of speed on the bearin


3/10

TILTING PAD THRUST BEARINGS: PERFORMANCE, AND IAdPROVEMENTS W I T H DIRECTED LUBRICATION 3

MEAN SLIDING SPEED-mls

Pad specific load, 35 bar.Oil flow, 45 litrelmin.

Fig. 13.3a. Effect of speed on bearing operating para-meters with constant pad specific loading

performance with a typical specific load of 35 bar*. At highsliding speeds the power absorbed increases nearly as thesquare of the speed. In addition to shearing losses beneaththe pads, the power absorbedalsoincludes drag losses on theexposed parts of the thrust collar and shaft. The draglosses (or churning losses) can amount to 50 per cent ormore of the total losses at high sliding speeds. There isalso a rapid increase with speed in maximum pad tempera-ture, which arises partly from the higher shearing rateand partly from the increased hot oil carry-over from onepad to the next. The high pad temperature represents areal limit to bearing loading (2), not only because thewhite-metal may lose its strength and begin to flowplastically but also for reasons concerned with thermalfatigue, sometimes known as thermal ratcheting (3).

The effect of pad specific loading at a constant highspeed (12 000 revlmin) is shown in Fig. 13.3b. Th e rela-tively slow increase in power absorbed with increasingload will be noted, showing that a considerable proportionof the power reduction is attributable to churning losses.It follows from Figs 1 3 . 3 ~and b that the rate of increase>f pad temperature due t o speed is of the same order asthat due to load. This has the practical effect of making it* 1 bar = 14.50 Ibf/in2 = lo5 N/m2.

Speed, 12 000 revlmin.Oil flow, 45 litrelmin.

Fig. 13.36. Eff ect of pad specific loading on bearingoperating parameters at constant speed

difficult to reduce maximum pad temperatures by using alarger bearing as any decrease in temperature due to lowerloading can be largely offset by the increase in meansliding speed at the thrust pad.

Effect of oil supply arrangementDifferent arrangements of oil inlet and outlet positions inconjunction with different oil flow patterns inside thebearing were studied to see how they affect the supply offresh cool oil to the pads. The merits of each arrangementwere judged by the measured pad temperatures. Theprincipal arrangements tested are shown in Fig. 13.4.The main oil flow path through the bearing was varied byusing a combination of different oil inlet and outletpositions in conjunction with full or partial recirculationslots in the carrier rings. From simple considerations ofoil flow it could be thought that the best arrangement isgiven by (V) where the total flow is supplied at the innerperiphery of the loaded face and discharged at the surgeface. On the other hand, t he worst arrangement wouldseem to be (111), in which the oil enters opposite the collarand has to flow against the collar's centrifugal gradient tolubricate the pads; furthermore, most of the oil could passaround the collar and be discharged at the top withoutever reaching the pad inlet edge.


4/10

M. K. B I E L E C AND A. J. L E O PA R D

+ O u t M a ~ nt h r u s ta l l c a r

..

, In ! In

Arrangement I

Arrangement I1

Iln 1In

A r r a n g e m e n t IP

(out

A r r a n g e m e n t P

A r r a n g e m e n t 91

Fig 13.4. Oil supply arrangements

T he chailpc in maximum pad tem perature versus speedis shown in Fig. 13.5 for each of the six arrangemen ts inFig. 13.4, and it will be seen that the different oil flowarrangements have only a marginal influence on maximu mpad temperatures. This was attributed to the extensivechurning and mixing of the oil within the casing whichwas induced by the rotating thrust collar, and by thismethod a fairly uniform oil bath temperature is main-tained irrespective of oil entry arrangements or oil flowinside the bearing.

Effect of oil quantity

The oil temperature rise through a t i l t ing pad bearing

assembly (i.e. the difference in housing oil inlet and outlettemperatures) is a parameter used extensively in designand operation.

Fig. 13.6 shows th e effect of oil quantity on the maxi-mum pad temperature of a bearing operating at 6000 and8000 rev/min with the corresponding temperature risemarked against each test point. I t will be seen tha t increas-ing the oil flow to give temperature rises of less than15-20 deg C has little beneficial effect on bearing per-formance as measured by pad surface temperature.Conversely, however, relatively large changes in pad

surface temperature can occur with small oil flows, i.e. athigh tem pera ture rises.

Effect of oil pressure

At low sliding speeds a bearing performs satisfactorily

when immersed in an unpressurized oil bath, but athigher speeds the centrifugal effects of the collar on thesurrounding oil become substantial. Recirculation slotsare provided a t the back of the carrier ring to equalize thepressure gradients created by the collar, but their successis only partial.

Tests were carried out at constant speeds to measure thestatic pressures at the outer and inner peripheries ofth e pads for given oil supp ly pressures. Fig. 13.7 shows theresults for 9000 rev/min, which was the maximum possibletest rig speed when the tests were performed. It will beseen that pressures at the outer periphery of the pads weresubstantially less tha n the supply pressure, and pressuresbelow atmospheric were measured at the inner periphery

of the pads with low oil supply pressures. This wasparticularly noticeable at high speeds where highertemperature readings indicated oil starvation at the padinlet. Th ese sub-atm ospheric pressures can be avoided byincreasing the supply pressure. Other workers(4) haveshown tha t the safety factor of a tilting pad thru st bearing

Pad specific load, 42 bar.Oil inlet temperature, 50C.Oil outlet temperature, 67C.

Fig. 13.5. Effect of oil flo w arrangements shown inFig. 13.4 on maximum pad temperature


5/10

TILTING PAD THRUST BEARINGS: PERFORMANCE, AND IMPROVEMENTS WITH DIRECTED LUBRICATION 5

is increased by raising the housing pressure to relativelyhigh values (10 bar), but from tests performed it isconcluded that supply pressures in the range of 1-1.5 barare sufficient for the majority of modern applications.

U

14 0v

I13 0

+uP:W

g 120WC

a

2 11 0z3

I5 ro o

Effect of misalignmentAbout 0.015 mm is a typical order of magnitude for the.ninimum film thickness separating the rotating collarfrom the pads. Thus, to ensure a reasonable load distribu-tion a very high degree of alignment is required betweenthe components concerned. Fundamentally there are threemain types of misalignment:

II

(1) differing pad heights;

(2) misaligned thrust collar with respect to thrust facein moving plane ('swash plate action');(3 ) misaligned pad carrier ring with respect to the

collar face in a static plane (a ) when initially assembledand (b ) when machine running.

I

The first two types can best be dealt with by propermachining and inspection. The third form, which in somecases can be quite large, is the most difficult to predict andcontrol. Several methods are used to ensure an alignedsupport, e.g. lever type equalizing rings, hydraulic pistons,springs, and spherical seats. Tests (5) have indicated thatthe lever type equalizing ring does not deal very effectivelywith changes in alignment of the bearing support,probably due to the high friction at the transfer points

between the levers. Hydraulic pistons give excellent per-formance but are expensive. With springs, the amount ofcompensation is limited by the permissible axial move-ment. The action of a spherical seat, whilst suitable foradjustment of alignment at assembly, has poor self-aligning properties under load owing to friction effects.

An ideal support should have high axial stiffness com--'-ined with freedom to tilt diametrically. Supporting the

caring on a fluid-filled flexible ring meets these require-ments, and such a device is shown in Fig. 13.8.

7*

O5 I 0 15 20 25 30 35 4045 50 100 150OIL FLOW- ~ /m i n

Pad specific load, 42 bar. Oil inlet temperature, 50C.

Fig. 13.6. Effect of oil flo w on maximum pad temperature

0 .3

-a

2 0.2-Ina

3 0. 1Ca

W

5 0InV)W(LP.

" -0.1-

8 1 1 1 I I I I I-0.2

(-1 Denotes pressures below atmospheric

Fig. 13.7. Effect o f oil supply pressure on static pressuresat inner and outer peripheries of thrust pad at 9000rev/min

Thru st pad Flexible

Fig. 13.8. Fluid filled alignment device


6/10

M. K. BIELEC AND A . J. LEOPARD

SHAFT S P E E D - r e v / m ~ n

Pad specific load, 35 bar.Oil flow, 2 3 litrelmin.M denotes alignment in m/m.

Fig. 13.98. Effect o f support misalignment at various

speeds on maximum pad temperature

Pad specific load, 35 bar.Oil flow, 23 litrelmin.M denotes alignment in mlm.

Fig. 1 3 . 9 ~ . Effec t of support misalignment on film

thickness

100; I I 1 I I I I I 10.0002 0.0004 0.0006 0.0008 0.0010 0-0012 0.0014 0.0016

MISALIGNMENT OF SUPPORT- m/rn

Speed, 10 000 revlmin. Oil flow, 2 3 litrelmin.

Fig. 13.9b. Effect of support misalignment on maximum pad temperature


7/10

TIL TIN G PAD THRU ST BEARINGS: PERFORMANCE, AND IMPROVEMENTS WIT H DIRECTED LUBRICATION 7

MISALIGNMENT OF SUPPORT -m/m

Pad specific load, 35 bar. Oil flow, 23 litrelmin.

Fig.13.9d.

Effect of support misalignment on minim um film thickness

A series of tests were carried out i n which a thrust ringwas supported on misaligned supports to determine theeffect of misal.ignment of type (3). In these tests themaximum misalignment tested was 0.0016 m/m and, inaddition to the normal Orion instrumentation, capacitanceprobes were fitted to the pad carrier ring to determine its:rue misalignment with respect to the thrust collar. It isof interest that at specific loads in excess of 15 bar, theamount of misalignment measured by these probes wasonly 30-50 per cent of tha t nominally imparted to thecarrier ring by its tapered support.

Some of the results from these tests are shown in Fig.13.9~-d. To simplify graphical presentation, only theresults on the plane of misalignment of the support for anominal bearing specific loading of 35 bar are shown.

From Figs 1 3 . 9 ~and b it can be seen that the pad surfacetemperature is scarcely affected by misalignment at lowspeed (1000 rev/min) but is considerably affected at highspeed (1 1 000 revlmin). I t is also noteworthy that the padtemperatures rise rapidly with initial small amounts ofmisalignment, after which further misalignment has anegligible effect.

Figs 13.9 ~and d show the effect of misalignment on theoil film thickness and it will be seen that both pad inletand outlet film thicknesses were proportionately reducedas misalignment was increased. I t was estimated that theheaviest loaded pad under maximum misalignment con-ditions had a specific pressure i n excess of 150 bar. I t isconsidered that the apparent discrepancy between tem-perature and film thickness trends with increasingmisalignment is caused by pad distortion.

The range of misalignments tested was consistent withthat which can occur in practice, and illustrates one of theprincipal reasons for bearings being underrated in current'ndustrial practice. When the thrust rings were supportedan the device shown in Fig. 13.8, the pad temperatures andfilm thicknesses of the pads in the ring were as uniform

under maximum misalignment conditions as those thatwere recorded with the standard thrust ring after mostcareful machining and assembly (shown as datum curvein Fig. 13.9~). A strong case can therefore be made forloading thrust bearings more heavily if they are fitted withan effective alignment device.

M E T H O D S O F R E D U C I N G P O W ER L O S S

It has been mentioned earlier that churning losses in high-speed tilting pad thrust assemblies can represent 50 percent or more of the total losses. For this reason various

attempts have previously been made to reduce these lossesby fitting seals to various parts of the assembly, and moreradical improvements have recently been made bychanging from the conventional 'fully flooded' method oflubrication.

Collar seals and 'L' seals

The commonest arrangement designed to reduce powerloss is the collar seal, whose object is to maintain an airpocket around the thrust collar periphery and thuseliminate collar periphery drag losses. As these losses canbe as high as 12 per cent of the total for a double thrustbearing, the use of these seals is attractive. Tests on theOrion rig, however, showed that fitting collar seals onlyresulted in a saving of about 3 per cent for the bearingoperating at 9000 rev/min with an oil supply pressure of1 bar. The reasons for this discrepancy between theoryand practice are principally (a) power loss on the seals and(b) difficulty in draining the annulus around the collarsufficiently well to ensure complete aeration, taking intoaccount the appreciable quantity of oil leaking throughboth seals.

Face type seals on the inner periphery of the thrust ring(sometimes called 'L' seals) have also been used toeliminate the drag losses on the periphery of the shaft.


8/10

8 1LI. K. BIELEC AND A. J. LEOPARD

Tests in the Orion rig failed to detect a power loss savingwith a bearing fitted with this device; however, thetheoretical saving for the case tested was only about 3 percent of the total, which was the same order as the overallrig accuracy at the time of testing.

With the collar seals and the 'L' seals it was noted thatthe pad surface temperature was slightly but consistentlyhigher (2-5 degC) than when the bearing was not fittedwith these devices. As the use of these seals can involvemore complicated oil supply systems and certain otherpractical disadvantages, their use can generally only bejustified in special designs which have unusual geometricproportions.

Directed lubrication

Following the relatively unpromising. results reportedabove on power loss reduction by means of seals, a morefundamental approach was felt to be required if anysignificant improvement was to be made. If 50 per cent or

more of the power absorbed is due to churning losses inhigh-speed bearings, it should be possible to eliminatethese completely if bearings could operate in a fullydrained housing with oil only present i n the areas betweenthe thrust pads and the collar, i.e. where the hydrodynamicoil films are formed. The lubricating oil has therefore to bedirected to the leading edge of every pad and removedfrom the trailing edge. In addition to saving power loss,this method should also improve bearing performance asit would eliminate hot oil carry-over f rom pad to pad.Such a system of lubrication is referred to as 'directed'lubrication.

The principal approaches were tried to meet the aboverequirements :

(1) Nozzles were placed at the inner periphery of thebearing directed radially outwards towards the leadingand trailing edges of the pads.

(2) Combined nozzle/scraper blocks were placedbetween each pair of pads. The nozzles were arrangedalong the whole radial width of the pads and weredirected towards leading and (in some cases) trailingedges of the pads.

(3) Conical nozzle systems were incorporated in thestop pins situated between each thrust pad. These arereferred to as umbrella sprayer nozzles.It may be noted that the first and third approaches use

jets of'oil to scour the hot oil carry-over from the collar,

while the second approach is designed to remove this oilby scraping.Initial investigation and development of these systems

was carried out on the Scylla rig. The principal criteriaused to evaluate the systems were maximum pad surfacetemperature and oil flow for given operating conditions;measurements of motor current enabled a rough assess-ment of power absorbed to be made. I t was found that theumbrella sprayer nozzles showed a marked superiorityover the other systems, especially at the highest speedsand loads tested (3000 rev/min = 100 m/s at 57 bar).

Following these initial results further tests were carriedout to optimize the design of the umbrella sprayer nozzlein terms of jet size, jet angle, number of jets per nozzle,jet velocity, and oil quantity. These tests emphasized theimportance of jet momentum in the removal of hot oilcarry-over. A typical thrust bearing fitted with umbrellasprayer nozzles is shown in Fig. 13.10.

A final series of tests were carried out on the Orion rigwhich enabled a direct comparison to be made betweenflooded and directed lubrication bearings with accuratemeasurements being made of power absorbed, pad tem-perature, film thickness, etc. The power available limitedthe maximum speed of these tests to 60 m/s (12 000rev/min) for flooded lubrication and 76 m/s (15 000rev/min) for directed lubrication.

Fig. 13.11~shows typical test results in which it will beseen that even at the relatively low speed of 60 m/s thedirected lubrication bearing achieved a reduction in powerloss of 45 per cent. A very practical demonstration of thiswas that the Orion rig was limited to 9000 rev/min whenall its thrust bearings were flood lubricated; after conver-sion of the test double thrust bearing and the reaction"thrust bearing to directed lubrica tion it was possible to runat 15 000 rev/min. T he curves in Fig. 13.1 lb for maximumpad temperature show that the bearing runs cooler withdirected lubrication at all speeds, and this differenceincreases with speed and becomes substantial (10-20degC) at high sliding speeds. It is interesting to note thatthe bearing with directed lubrication can support 14 batgreater pad specific loading for the same maximum pad

Fig. 13.10. 240 mm 0.d. tilting pad thrust bearingfitte d for directed lubrication


9/10

I'ILTING PAD THRUST BEARINGS: PERFORMANCE, AND IMPKOVEMENTS WITH DIRECTED LUBRICATION 9

S H A F T SPEED -rev/min

Flooded lubrication.- - - - Directed lubrication.Oil flow, 45 litrelmin.

Fig. 13.11a. Comparison of powe r absorbed by floodedand directed lubric ation

~mperature at a sliding speed of 60 m/s. This is animportant finding, since in high-speed applications thepad specific load is frequently reduced to keep thepad temperature low. Therefore, the use of directedlubrication permits higher specific loading, thus requiringa smaller bearing to support a given load; this in turn leadsto further saving in power due to a reduced bearingdiameter. Elimination of the parasitic churning losses incombination with a smaller bearing selection can lead to70 per cent or more reduction in the power absorbed by adouble thrust bearing required to support a given load athigh (100 m/s) sliding speeds.

Although the minimum film thickness is not a limitingcriterion in high-speed tilting pad thrust bearings, thecurves for the two systems of lubrication (Fig. 13.1 1c) showthat the bearing with directed lubrication operates with asubstantially thicker oil film, because cooler oil is supplieddirect to the pad inlet.

In the event of the loss of oil supply pressure, a floodlubricated tilting pad thrust bearing would at first sight

-. appear to have an inbuilt advantage over a directedlubrication bearing as it has an oil reservoir in the casing.However, it should be noted that in high-speed bearings

a depressurized casing will result in a fairly rapid centri-fuging of the oil away from the pads.

Fig. 13.12 shows some results from a series of testsdesigned to assess the sensitivity of directed lubrication toa reduction in oil supply pressure. Tests were also carriedout on the effect of a complete failure of the oil supply.

These showed that with a specific load of 35 bar at 3000rev/min no damage to the pads had occurred 30 s after oilwas cut off; at 9000 rev/min, however, wiping of thebearing surface commenced 13 s after oil was cut off.I t is felt tha t these results are comparable with the effectof oil supply failure in high-speed journal bearings.

C O N C L U S I O N S

The following main conclusions may be drawn from thework described above:

(1) Th e use of directed lubrication in high-speed thrustassemblies achieves substantial reductions in powerabsorbed, at the same time reducing the pad surfacetemperature and increasing the film thickness.

(2) Th e use of seals in high-speed thrust assemblies toreduce power absorbed is of only small value.

(3) The importance of accurate alignment has beendemonstrated, and if considerable overloading of some


Fig. 13.11b. Comparison of maximum pad temperaturewith flooded and directed lubrication


10/10

10 M. K. BIELEC A N D A. J. L E O PA R D


F ig. 13 .1 1~ . Comparison of minimum fi lm thicknesswi th flooded and directed lubrication

pads is not to result, with consequent reduction i n bearingsafety factor, it is necessary to achieve extremely highstandards. I t is possible to reduce the size of a bearing fora given load if an efficient alignment device is used.

(4) The use of relatively complicated internal andexternal oil flow arrangements for flood lubricated bearingscan only adhieve a limited improvement in pad surfacetemperature; at best, about 7 degC.

(5) Increasing the oil flow through the assembly is onlyof value if temperature rises in excess of about 20 degCare present.

(6) Sufficient oil pressure above ambient should bepresent in the housing of a flood lubricated bearing toavoid starvation of the oil supply to the pads due tocentrifugal effects. For most applications, 1 - 1 5 bar issufficient.

- - a

OIL SUPPLY PRESSURE BAR

Pad specific load, 35 bar.

Fig. 13.12. Effect o f reduced oil supply pressure wi thdirected lubrication on maximum pad temperature

A C K N O W L E D G E M E N T S

The authors would like to thank the Glacier Metal Co.Ltd for permission to publish this paper, and express theirgratitude to all who assisted in its preparation. Acknow-ledgement is also made to the Ministry of Defence (D.G.Ships) for sponsoring part of the work on the Orion rig.Some features of directed lubrication described in thispaper are covered by Glacier Metal Co. Ltd G.B. PatentNo. 1 120 624, G.B. Patent Application No. 42 951166and corresponding foreign patents and applications.

A P P E N D I X 13.1R E F E R E N C E S

(I ) BUTCHER, A. E. 'Developments in the measurement of oil-film thickness in dynamically loaded bearings', S y m p .Experimental Methods in Tribology, Proc. Znstn mech. Engrs1967-68 182 (Pt 3G), 105.

(2 ) MARTIN, F. A. 'Tilting pad thrust bearings: rapid designaids', Paper 16 at this Convention.

(3) BOAS, W. and HONEYCOMBE, R. W. K. 'The deformation oftin base bearing alloys by heating and cooling', J . Znst.Meta ls 1947, 433.

(4) TRIFONOV,E. V. and YAMPOLSKII, S. L. 'Effects of oilpressure on the load-carrying capacity of steam turbinethrust bearings', ~nergomashinot roenie1957 1, 8 .

( 5 ) GUSTAFSON, R. E. 'Behaviour of a pivoted-pad thrust bearing

during start-up', A.S.M.E. Paper 65-Lub-S-13,1965 (June).

tilting_pad_thrust_bearings_performance_directed_lubrication

Documents