design and development of experimental setup for …psrcentre.org/images/extraimages/6 314014.pdfan...

Abstract— An experimental setup is designed to test water

lubricated bearing simulating the conditions of its practical

application. The bearing that is being used for the study is the

commercially available water lubricated bearing which has a flexible

rubber bonded with a rigid metal shell. The bearing has 8 lands and 8

axial grooves throughout the length of the bearing. The lubricant

enters bearing axially through a pressurized inlet tank and exits to

outlet tank which is at sufficiently low pressure. A variable speed DC

motor is connected via a pulley arrangement to drive the journal

bearing assembly. The load on the bearing is applied through the

dead weight system which acts both in upward and downward

direction so that a net load is applied on the bearing. The pressure

tapping in the circumferential direction in the central plane of the

bearing are used to find the hydrodynamic pressure generated. The

axial pressure change along the length of the bearing is also

established. To calculate the eccentricity, 2 pairs of proximity sensors

which are perpendicular to each other are mounted at the two ends of

the test bearing housing. The inlet and outlet temperature of the

lubricant is also found from the thermocouple attached to the two

supply tanks.

Keywords— hydrodynamic pressure, needle bearing, staves,

water lubricated bearing.

I. INTRODUCTION

EARINGS are one of the important machine elements in

any equipment. The design of journal bearings is

considered important to the development of rotating

machinery. The increasing ecological awareness, more

stringent requirements for environmental protection have

resulted in ship designers, manufacturers searching for

inexpensive, simple and reliable design solutions for building

new ships and modernizing existing ones [1]. Charles

Sherwood is generally credited with the accidental discovery

of water-lubricated rubber journal bearings in the 1920s [2].

The earlier water lubricated bearings were made up of lignum

vitae, phenolic compound or rubber [3]. The bearings that are

being used today use different types of polymer or hard rubber

Ravindra Mallya is a research scholar in Mechanical Engineering

Department, Manipal Institute of Technology, Manipal University,

Manipal-576104, INDIA. (e-mail: [email protected]).

Satish Shenoy B is Professor in the Aeronautical and Automobile

Engineering Department, Manipal Institute of Technology, Manipal

University, Manipal – 576104, INDIA. (Phone: +91-0820-2925472; e-mail:

[email protected]).

Raghuvir Pai. B is with the Mechanical Engineering Department, Manipal

Institute of Technology, Manipal University, Manipal – 576104, INDIA.

(Phone: +91-0820-2925465; e-mail: [email protected]).

based on the requirement of the customer and the working

environment. Rubber and water make the perfect relationship

for a bearing. The natural resilience of rubber gives the

bearing its shock, vibration and noise absorption properties.

The most popular design used today is the straight fluted

bearing. This design of bearing consists of a number of load

carrying lands or staves, separated by flutes oriented axially in

the bearing. (Fig 1)

Fig. 1 Straight fluted water lubricated bearing [2]

A water lubricated bearing consists of a cylindrical shell and

liner, made from hard plastics from a casting process and

rubber. The flutes supply the bearing with lubricant, which

enters at one end of the bearing and leaves at the other. The

water enters the bearing through the longitudinal grooves and

moves radically between the propeller shaft and the bearing

face in a thin film. Once this film, or wedge, has developed the

shaft does not actually come into contact with the bearing. The

hydrodynamic water wedge is formed between the shaft and

bearing, even at very low shaft speed. These types of bearings

find application in water power plants and pumps in which a

pumped liquid, usually water, serves as a lubricating medium.

They are also required in mining industry, water conditioning

stations or land melioration applications, Navy ships, boats

and submarines. Cabrera et al; [2] have developed a test rig to

find the film pressure distribution in the water lubricated

rubber journal bearing. The measurements of pressure

indicated that the film pressure profiles are very different from

those of the conventional rigid bearings. The behavior of the

bearings was theoretically investigated using computational

fluid dynamics and compared with the experimental values

conducted on the test rig designed. Litwin [4] has developed

an experimental setup to test the propeller shaft bearings of the

Design and Development of Experimental Setup

for Testing Water Lubricated Bearing

Ravindra Mallya, Satish Shenoy B, and Raghuvir Pai B

B

3rd International Conference on Mechanical, Electronics and Mechatronics Engineering (ICMEME'2014) March 19-20, 2014 Abu Dhabi (UAE)

22

ship. He has tested the water lubricated bearings for its

hydrodynamic pressure developed in the water film. The

pressure measuring system was a pressure sensor mounted

inside the shaft and the signals were transmitted using a

wireless system. The trajectory of the shaft was determined by

having two pairs of non contact transducers with an accuracy

of 1µm on either side of the shaft. The experiments were

conducted for static as well as dynamic loads. Pai [5] et al

have compared the experimental results with the fluid flow

modeled in CFD for a journal bearing with three equi-spaced

axial grooves and supplied with water from one end of the

bearing. A bearing test rig was developed to evaluate the non –

metallic journal bearing performance when lubricated by

water. Their objective of study was to predict the pressure in

the lubricant (water), measure the pressure in the lubricant

(water) film in the journal bearing operating with various

conditions.

There was a difference in the predicted pressure contours

and the experimentally measured contours. Axial pressure

variations along the groove agreed with experimental trends

from a qualitative perspective, but did not equal from

quantitative perspective. Relatively little experimental work

has been carried out previously in understanding the

performance characteristics of these bearings. Wang [6] et al

have also investigated the film pressure distribution of water

lubricated bearing based on wireless measurement technique.

Electromagnetic loading and wireless measurement of the film

pressure have been incorporated in their study. The lubricant

water is supplied to the bearing through the 6 inlet ports at one

end and exits the bearing from similar outlet ports at the other

end.

A simulation model of clearance between shaft and the

bearing was also established using ProE and GAMBIT, and

the simulation results were obtained by ANSYS and FLUENT.

In the present study an experimental set up is developed to test

the commercially available water lubricated bearings. The

objective of the study was to simulate the practical conditions

of the bearing and study the axial and hydrodynamic pressure

behavior as the lubricant flows through the length of the

bearing. The test setup has the capability to determine the

static performance parameters such as eccentricity, pressure

developed in the film, load carrying capacity, flow rate,

temperature change were also investigated for different speeds

and loads.

II. EXPERIMENTAL TEST SETUP

The assembled view of the test setup for water lubricated

bearing is shown in Fig. 2. The experimental set up uses the

commercially available water lubricated bearing with 8

grooves and 8 lands. The bearing is made of hard rubber

molded inside a gun metal liner. The bore diameter of the test

bearing is approximately 64 mm and length of the bearing is

250 mm. It is press fitted inside a bearing housing made of

mild steel. A mild steel hollow shaft of 750 mm in length and

63 mm in outside diameter and 25 mm inside diameter is used

as the drive shaft.

Fig. 2 Water lubricated bearing test setup

The drive shaft is driven via a step-down pulley

arrangement connected to a 5 hp DC Motor (1500 rpm). The

operation speed range of the drive shaft is from 0 to 530 rpm.

The shaft is ground finished and the diametral clearance

between the test bearing and the shaft is 0.83 mm. The

clearance is sufficiently large when compared with oil

lubricated bearing. This is to account for the absorption of

water by the rubber bearing and thus causing it to swell. If the

clearance is less, there is a possibility of bearing clenching the

shaft or bearing blocking, which will be difficult to

disassemble.

A. Test Bearing

Fig. 3 Assembled view of test bearing inside the housing

The assembled view of the test bearing into the housing is

shown in Fig. 3. The test bearing is the locally available rubber

bearing with eight lands and eight grooves which are

throughout the length of the bearing. The bearing is made up

of hard rubber molded in a gun metal liner. The average inner

diameter of the bearing measured using the Coordinate


23

Measuring Machine of Mitotoyu make at the land or staves is

64.035 mm. The length of the bearing is 254 mm. The L/D

ratio is approximately 4. The test bearing is then press fitted

inside a mild steel housing. 10 holes of 5 mm diameter are

drilled at regular intervals in the central plane along the

circumferential direction where the hydrodynamic pressure

tapping are to be connected. Holes are also drilled in the axial

direction of the bearing to measure the axial pressure drop

along the length of the bearing.

B. Loading System

Fig. 4 Assembled view of cage and test bearing

Fig. 4 shows assembly of test bearing with the loading cage.

The test bearing is loaded using a cage type arrangement to

apply the dead weights which acts in both upward and

downward direction. 6 deep groove ball bearings of 10 mm ID

are fastened at 120o angle in circumferential direction to two

annular disc made of mild steel of 10 mm thickness. There is

an option provided to move one of the bearing for making

adjustment during assembly by cutting a slot on the disc. The

two annular discs of 8 mm are fastened to a tie rod of 15 mm

diameter. Grooves are cut on the tie rod for clasping the wire

rope. The whole test bearing with the housing is inserted into

the cage. The test bearing rests on the 6 ball bearings similar to

a steady rest. The cage is fabricated in such a way so as to

keep the two tie rods in the vertical plane on either side of the

test bearing. The tie rods on the top and bottom of the bearing

are attached to a hook and wire rope to dead weight loading

system. A dial type gauge of 100 kg capacity is mounted in

between the tie rod and the wire rope.

Load on the bearing is applied both in upward and

downward direction. Fig. 5 shows a schematic diagram of

loading of the test bearing. The net load acting on the test

bearing is shown by the dial gauge. This is done in order to

locate the journal centre in the hot rotating condition and to

measure the eccentricity ratio.

Fig. 5 Schematic diagram of loading system

The important feature of the experimental set up is the

bearing centre is located seperately for each load or speed

warm-up setting. The loading mechanism allows us to apply

many pairs of different vertical loads in the upwards and

downward directions. For each pair, the horizontal and vertical

movements of the bearing can be measured using the readings

from the display.

C. Drive Shaft

The drive shaft is a mild steel hollow shaft which is

approximately 750 mm in length. The outside diameter of the

drive shaft is ground to 63.206 mm. The inner diameter of the

shaft is approximately 25 mm. The diametric clearance

between the bearing and the shaft is approximately 0.83 mm.

A provision is made for mounting a flush type pressure

transducer at the centre of the shaft, which will measure the

steady state hydrodynamic pressure generated. The drive shaft

is supported by rolling contact bearings on either side. The

drive shaft is run using a pulley and belt driven by variable

speed DC Motor.

D. Drive Shaft Design of Drive Shaft

The power rating of the variable speed DC Motor is 3.75

kW at 1500 rpm. The variable speed DC Motor is controlled

using a speed controller and the maximum speed at the Motor

shaft is around 750 rpm. A 101.6 mm (4”) diameter pulley is

mounted on the Motor shaft and connected via a belt drive to

152.4 mm (6”) diameter pulley on the drive shaft. The step

down pulley system runs the shaft at a maximum speed of

approximately 530 rpm. The power transmitted to the shaft at

530 rpm and torque of 47.75 Nm is calculated to be 2.65kW.

The test bearing, the drive pulley, the load applied on the

bearing act on the support bearings at the two ends through the

drive shaft. The shaft has to be designed for its bending and

shear strength. The approximate weights of the test bearing,

bearing housing, self weight of the shaft, and the load to be

applied on the test bearing all together is found to be 1373 N.

The belt tension forces acting in the vertical plane on the

pulley and thereby acting on the shaft are the additional loads.


24

The total load at the location of the pulley is the total tension

and the self-weight of the pulley which comes to 926.5 N. The

total load of the bearing, housing and the load to be applied on

the test bearing is found to be 1524 N, which can be

considered as a uniformly distributed load along the bearing

length of 254 mm.

The Free Body Diagram of the drive shaft is shown in

Fig. 6. The support bearings are at the location B and E. The

drive pulley is at the location F and the test bearing between C

and D. The test bearing load on the shaft is considered as a

uniformly distributed load as the load acts throughout the

bearing length. The maximum bending moment is developed at

the support bearing E = 120.45 kN-mm.

Fig. 6 Free body diagram of drive shaft

The hollow shaft is ground to an OD of 63 mm, the ID of

the shaft 25 mm. Using the shear stress theory for hollow shaft

design, 2

max 2.7 N / mm . The maximum shear stress

developed in the shaft after taking a factor of safety of 3 is

much below the permissible value, making the shaft safe for

use.

E. Design of Support Bearings

The hollow shaft is supported by deep groove ball bearings

on one side and needle roller bearing on the other side. The

needle roller bearing helps in easy disassembly of the test

setup. The bearing mounted on the bearing supports is

assumed to be subjected to axial force of 1000 N. The radial

forces acting on the bearing is approximately 2450 N. Taking

the radial load as thrice its value, for a fairly considerate

design, the radial load is calculated as 7350 N. The bearing is

to be mounted on the shaft of outside diameter of 55 mm.

Selecting the values of X = 1 and Y = 0, from the catalogue of

SKF bearings for single row deep groove ball bearings, the

equivalent dynamic load is calculated as P = 7.35 kN.

Assuming the bearing to work for 10,000 hr of life running

at 1500 rpm. Using the L10 equation to calculate life, the life of

ball bearing comes to 900 million revolutions. The dynamic

load capacity of the ball bearing C is 70.96 kN.

From the catalogue of SKF bearings the 6311-2Z bearing is

selected. The bearing is a single groove ball bearing which has

an ID of 55 mm and OD of 120 mm. The width of the bearing

is 29 mm and it is shielded on both sides. Similarly,

calculating life for Needle roller bearings. These bearings are

not designed for taking axial loads, i.e. Fa = 0. The selected

bearing from the list is NK 55/25. The ID of the bearing is 55

mm and the OD is 68 mm. The width of the bearing is 35 mm.

The L10 life of the bearing is 650 million revolutions and the

dynamic load capacity found is 52.3 kN. The needle bearing is

press fitted inside the bearing housing. This bearing has

hardened rollers which are located inside a metal cage and run

on the surface of the shaft during its operation. These needle

type bearings help in easy dismantling of the test setup without

much time consumption.

F. Lubrication System

The lubricant water is pumped from a 200 litre storage tank

into the inlet tank adjacent to the test bearing using a

centrifugal pump of capacity 0.5 hp and 30 m head. A control

valve is provided at the pump outlet to vary the pressure of the

lubricant. The extra lubricant returns to the storage tank

through a back pressure pipe. The tank is fabricated using a

metal pipe welded with mild steel sheet on both sides.

Provision is made for placing lip seal for preventing the

leakage of the lubricant as the drive shaft passes through both

the tank. The lubricant is fed into the inlet tank through

flexible tubes. The lubricant fills the inlet tank and enters into

the bearing through the groove and clearance between the

drive shaft and the test bearing. The water exits from the other

end of the bearing into the outlet tank and back to the storage

tank thus creating a closed loop system of lubrication. 2 dial

type pressure gauges are mounted on both the tanks to monitor

the inlet and outlet pressure of the lubricant. Thermocouples

with capacity to measure temperature up to 50oC are attached

to the tanks; the display panel connected to the thermocouple

will give the temperature change if any of the lubricant.

G. Design of Inlet/Outlet Tanks

The expected pressure inside the inlet/outlet tank is 50 kPa.

The diameter of the tank is 152 mm and the width 65 mm. The

thickness of the metal sheet used in fabrication is 2.5 mm.

Using the design equations for thin walled cylinders, the

tangential stress developed in the tank

is 2

t 2.94 N / mm . The longitudinal stress developed in the

tank is 2

l 0.735N / mm . Both the stresses are much below

the permissible stresses of mild steel making the tank safe for

operation.

III. INSTRUMENTATION

To measure the hydrodynamic pressure developed in the

bearing, a flush type diaphragm pressure transducer is

mounted on the shaft. The pressure sensor is fastened to a

fixture which is then fastened on to the drive shaft such that

the sensing area is slightly below the surface of the shaft and

only the sensor top surface is exposed to the pressure

developed. A schematic diagram of pressure sensor mounted

on the shaft is shown in the Fig. 7. The leads from the pressure

transducer are connected via a slip ring to the data acquisition

system.

The pressure transducer is of Honeywell make and

maximum pressure that can be measured is 20 bar (300 psi).

An inline amplifier was connected between the sensor and the

data acquisition board to amplify the output signal from the

sensor.


25

Fig. 7 Schematic diagram of pressure sensor mounted on

the drive shaft

The test bearing is attached with two pairs of non-contact

type displacement sensor (NC-DVRT-2.5) of Microstrain

make. The range of these sensors is 0 – 2.5 mm and they are

submersible type sensors. These sensors were calibrated by the

manfucturer to the required interacting surface. A circuit was

built (Fig. 8) in the data acquisition software QUARC using

certain functions of the software and all the four proximity

sensors were connected to the power supply and data

acquisition board.

The sensing probe is connected via a DEMOD- DC, a signal

conditioning connector to a DC power supply. The output of

all the 4 sensors was displayed simultaneously on the computer

connected to the data acquisition board. The proximity sensors

can be used to measure the eccentricity and misalignement of

the test bearing. The data acquisition system uses Q8 board of

Quensar make which interfaces the proximity sensors and

pressure sensor through a software Quarc. A simple circuit is

designed in the software using the built in functions. The

Quarc software interfaces with MATLAB software to give the

output of all the sensors simultaneously which can be

monitored as the experiment is working or can be stored as

data file in the computer.

Fig. 8 A circuit connecting 4 sensors to DAQ board

There was suffcient amount of leakage of water between the

test bearing surface and the adjacent inlet tank surface. The

leakage of water had to be prevented so that the bearing is

supplied with sufficient lubricant. The issue was sorted out by

modifying the design of the inlet and the outlet tanks. The

cylindrical tanks are attached to the test bearing as a single

member, thus preventing any possibilty of leakage.

IV. CONCLUSIONS

The design and development of a test stand to simulate the

practical conditions of water lubricated bearings is complete.

The water lubricated bearings are tested for different speeds

and loads to know the steady state characteristics such as

eccentricity, hydrodynamic pressure, load carrying capacity

and flow rate. Currently tests are being conducted. The

experience gained will enable us to design better rig for

measuring the performance of water lubricated bearings.

REFERENCES

1. Wojciech Litwin, 2009, “Water lubricated bearings of ship propeller

shafts – problems, experimental tests and theoretical investigations”,

Polish Maritime Research, 4(62) Vol. 16; pp 42 – 50.

2. D L Cabrera, N H Woolley, D R Allanson, Y D Tridimas, 2005, “Film

pressure distribution in water lubricated rubber journal bearings”, Proc.

IMechE, Journal of Engineering Tribology, Vol. 219, pp 125 – 132.

3. Nakano et al, July 27, 1976, United States Patent, Patent Number –

3971606.

4. W Litwin, 2006, “The test stands for water lubricated marine main shaft

bearings”, NT 2006 – 5 – 103.

5. R Pai, D J Hargreaves, R Brown, 2001, “Modeling of fluid flow in a 3-

Axial groove water bearing using computational fluid dynamics”, 14th

Australian Fluid Mechanics Conference, Adelaide University,

Adelaide, Australia, pp 331 – 334.

6. Nan Wang, Qingfeng Meng, Pengpeng Wang, Tao Geng and Xiaoyang

Yuan, 2013, “Research on film pressure distribution of water lubricated

rubber bearing with multi axial grooves based on wireless measurement

technique”, ASME, J. Fluids Eng. 135(8), pp 084501-1 – 084501-6.

7. Raghuvir Pai, David W Parkins, 1998, “Measurement of Eccentricity

and attitude angle of a journal bearing”, 3rd National Conference on

Fluid Machinery, pp 181 – 188.

8. V B Bhandari, 2009, Design of Machine elements, TMH Publication.

9. K. Mahadevan, K. Balveera Reddy, 2013, Design Data Hand Book for

Mechanical Engineers, CBS Publication.

Ravindra Mallya was born in Mangalore on June 7th in the year 1980.

He has graduated in Mechanical Engineering in 2002 from Manipal Institute

of Technology, Manipal University, Manipal. He has completed his Masters

Degree in Design Engineering, from Jawaharlal Nehru National College of

Engineering, Shimoga under the Visweswariah Technological University,

Karnataka in 2009. He has joined Canara Engineering College,

Benjanapadavu, as Lecturer in Department of Mechanical Engineering in

2005 and currently pursuing his full time PhD course in the area of Tribology

from Manipal University. His areas of interest are Water Lubricated Bearings,

Machine Design.

Satish Shenoy B was born and brought up in Someswara, Mangalore. He

finished his graduated in Mechanical Engineering in the year 1998 from

Mysore University. He completed his M.Tech in Production Engineering and

Systems Technology in 2001 from National Institute of Engineering, Mysore.

He obtained his Doctoral Degree in 2009 from Manipal University in

Tribology.

Shenoy is presently working as Professor in Department of Aeronautical and

Automobile Engineering, Manipal Institute of Technology, Manipal

University. His core expertise is in the domain of Computational Fluid

Dynamics, Applied Numerical Methods, Aircraft Structures, Finite Element

Methods, and Design of Manufacturing Tools. He has been the University

topper during his PG entrance examination. He has more than 25 journal

publication in the area of tribology to his credit.

Raghuvir Pai. B born in Ballambat on 26/03/1964 had his Bachelors in

Mechanical Engineering in 1982 from the University of Mysore. In 1988, he

went to the Indian Institute of Technology, Kharagpur and completed his

Doctoral degree in Tribology. From February 1995 to February 1996, he had

worked as a Department of Science and Technology (Government of India),

BOYSCAST Postdoctoral research fellow at Cranfield University, England.


26

He has research and teaching experience of 2 years (2000-2002) at

Queensland University of Technology, Brisbane, Australia.

Dr. Pai has published more than 100 research papers in Journals, International

and National conferences. He has supervised 04 Ph.D. students in the area of

Tribology Water Lubricated Bearings, Externally adjustable bearings and Tri-

taper bearings and Tribology of machining metal matrix composites. He was

a principal investigator for research projects by Philips, Bharat Heavy

Electricals and GE JFWTC Global Research Centre, Bangalore. He has

conducted more than 10 short courses in the field of Tribology. Currently he

is heading Director-Research (Technical), Manipal University, Manipal.


27

design and development of experimental setup for …psrcentre.org/images/extraimages/6 314014.pdfan...

Documents