Experiments on Ultrasonic Lubrication Using a Piezoelectrically-assisted Tribometer and Optical Profilometer

Summary

We present a protocol for using a piezoelectrically-assisted tribometer and optical profilometer to investigate the dependence of ultrasonic wear and friction reduction on linear velocity, contact pressure, and surface properties.

Abstract

Friction and wear are detrimental to engineered systems. Ultrasonic lubrication is achieved when the interface between two sliding surfaces is vibrated at a frequency above the acoustic range (20 kHz). As a solid-state technology, ultrasonic lubrication can be used where conventional lubricants are unfeasible or undesirable. Further, ultrasonic lubrication allows for electrical modulation of the effective friction coefficient between two sliding surfaces. This property enables adaptive systems that modify their frictional state and associated dynamic response as the operating conditions change. Surface wear can also be reduced through ultrasonic lubrication. We developed a protocol to investigate the dependence of friction force reduction and wear reduction on the linear sliding velocity between ultrasonically lubricated surfaces. A pin-on-disc tribometer was built which differs from commercial units in that a piezoelectric stack is used to vibrate the pin at 22 kHz normal to the rotating disc surface. Friction and wear metrics including effective friction force, volume loss, and surface roughness are measured without and with ultrasonic vibrations at a constant pressure of 1 to 4 MPa and three different sliding velocities: 20.3, 40.6, and 87 mm/sec. An optical profilometer is utilized to characterize the wear surfaces. The effective friction force is reduced by 62% at 20.3 mm/sec. Consistently with existing theories for ultrasonic lubrication, the percent reduction in friction force diminishes with increasing speed, down to 29% friction force reduction at 87 mm/sec. Wear reduction remains essentially constant (49%) at the three speeds considered.

Introduction

Friction exists at the interface of two contacting surfaces when they slide or roll relative to each other. Friction usually occurs along with abrasive or adhesive wear.¹ Ultrasonics is the science behind high frequency phenomena, that is, waves traveling at frequencies above the acoustic range (20 kHz). The field of ultrasonics encompasses two fundamentally different regimes. One regime involves low intensity waves like those utilized in imaging processes such as medical ultrasound or non-destructive inspection of structures. The other is a high power regime in which high-energy waves are utilized to execute or assist engineering processes such as welding of plastics and metals. It has been shown that application of the latter kind of ultrasonic vibrations at the interface of two surfaces in sliding contact reduces the effective friction force at the interface. This phenomenon is known as ultrasonic lubrication.

To achieve ultrasonic lubrication between two sliding objects, relative vibration at ultrasonic frequencies must be established between them. Vibrations are typically applied to one of the two objects, either in the longitudinal, transverse, or perpendicular direction relative to the sliding velocity. In this study, a tribometer’s pin is fitted with a piezoelectric actuator so that its tip vibrates in the direction perpendicular to the tribometer’s rotating disc. Piezoelectric materials are a class of “smart” materials that deform when exposed to electric fields, vibrating at the same frequency as the excitation field. Piezoelectric materials can vibrate at frequencies well into the MHz range. Being superimposed to the macroscopic velocity, ultrasonic vibrations have the effect of alternating the direction of the instantaneous friction force and the contact between the surfaces, which in combination leads to a reduction of the effective friction force and surface wear.

Ultrasonic friction reduction has been demonstrated in practical manufacturing systems. For example, this technology has been utilized to reduce the force between tool and work piece in metal machining and forming processes such as drilling, pressing, sheet rolling, and wire drawing. Benefits include improved surface finish² and a reduced need for expensive and environmentally harmful detergents to remove lubricants from the final product. There are potential applications of ultrasonic lubrication in other areas as well. For example, ultrasonic lubrication can substantially enhance the user experience in personal health care products by eliminating the need for lubricants or coatings. In automobile applications, friction modulation can improve the performance of ball joints whereas friction reduction between vehicle seats and rails facilitates seat movement, saving space and mass that would otherwise be occupied by traditional components and mechanisms. Ultrasonic lubrication can also help to improve fuel efficiency by reducing friction in powertrain and suspension systems.³ In space applications, where traditional lubricants cannot be used, ultrasonic lubrication can be employed to reduce wear and dramatically extend the life of critical components.

Laboratory demonstrations of friction reduction through ultrasonic lubrication are numerous. Friction reduction is quantified as the difference between the friction force measured without ultrasonic lubrication and the friction force with ultrasonic vibrations applied. In either case, the friction force is directly measured with force sensors. Littmann et al. ^4-5 connected a piezoelectrically-driven actuator to a slider, on which a force sensor and a frame were installed for measuring friction forces and applying normal loads. A pneumatic actuator was employed to push the slider together with the actuator along a guide rail. Ultrasonic vibrations were applied in the direction longitudinal to the sliding velocity. Bharadwaj and Dapino^6-7 conducted similar experiments using a piezoelectric stack actuator connected to a conical waveguide at either end of the stack. Contacts took place between the spherical edges of the cones and the surface of the guide rail. The effects of system parameters such as contact stiffness, normal load, and global stiffness were studied. Kumar and Hutchings⁸ installed a pin on a sonotrode which was energized by an ultrasonic transducer. Ultrasonic vibrations were generated and transmitted to the pin, which was placed in contact with a tool steel surface. Normal force was applied by a pneumatic cylinder and measured by a load cell. The relative motion between the pin and the disc was created by a reciprocating table.

Pohlman and Lehfeldt⁹ also implemented a pin-on-disc experiment. Unlike other studies, they employed a magnetostrictive transducer to generate ultrasonic vibrations. To study the optimum direction for ultrasonic friction reduction, the transducer was carefully aligned so that the vibrational direction was longitudinal, transverse, and vertical to the macroscopic velocity. They studied ultrasonic friction reduction on both dry and lubricated surfaces. Popov et al.¹⁰ utilized an actuator with conical waveguides. The actuator was placed in contact with a rotating base plate. Cones made of nine materials with various hardnesses were adopted to study the influence of material hardness on ultrasonic friction reduction. Dong and Dapino^11-13 used a piezoelectric transducer to generate and transmit ultrasonic vibrations to a prismatic waveguide with rounded edges. The longitudinal vibration causes vertical vibration due to Poisson’s effect. A slider with a curved top was placed under and in contact the waveguide. A frame was built to apply normal forces at the contact interface. The slider was pulled manually around the center area of the waveguide; the friction force was measured by a load cell that was connected to the slider.

Ultrasonically-induced wear reduction was also investigated and demonstrated. Volume loss, weight loss, and surface roughness changes are employed to quantify the severity of wear.Chowdhury and Helali¹⁴ vibrated a rotating disc in a pin-on-disc setup. The vibrations were generated by a support structure of two parallel plates located under the rotating disc. The top plate has a spherical ball installed off-center on the bottom surface, which slides in a slot that was engraved at the top surface of the bottom plate. The slot was machined with a periodically variable depth so that the top plate moves vertically during rotation. The frequencies ranged around 100 Hz according to the rotational speed.

Bryant and York^15-16 studied the effect of micro-vibrations on wear reduction. They inserted a carbon cylinder through a holder with one end rested on a spinning steel disc and the other end connected to a coil spring. In one case, the cylinder was snug fitted in the holder so that there was no space for vibration. In other cases, clearances were left to allow micro-vibrations of the cylinder while the cylinder was in contact with the spinning disc. The weight loss of the cylinder was measured to calculate the wear rate. It was shown that the self-generated micro-vibrations helped reduce wear by up to 50%.

Goto and Ashida^17-18 also adopted a pin-on-disc experiment. They connected pin samples with a transducer via a tapered cone and a horn. The pin vibrated in the direction perpendicular to the disc surface. A mass was connected to the transducer on its top for applying normal loads. Friction forces were translated from the torque that was applied to rotate the disc. Wear was identified as adhesive because both pin and disc were made of carbon steel. Wear rates were calculated from volume loss measurements.

It has been shown that linear speed plays an important role in ultrasonic lubrication. The experimental component of this research focuses on the dependence of friction and wear reduction on linear speed.

Protocol

1. Development of the Modified Tribometer

1. Install chuck-motor subsystem.
   1. Level vibration isolation table. Place DC motor on the table; level the motor with shims and fix it with struts and bolts. Place support frame around the motor.
   2. Connect splined shaft to the motor shaft using a key. Put support plate on the frame with the splined shaft going through the hole in the plate. Set thrust needle-roller bearing on the support plate and around the splined shaft. Lubricate the bearing with cutting fluids.
   3. Connect the splined shaft to the chuck through an adapter plate, which has a splined shaft coupling on one side and the chuck’s bolt pattern on the other side. At this point, the chuck is supported by the frame through the thrust bearing and connected to the motor through the adapter plate.
2. Install the gymbal assembly.
   1. Build the support frame using U-channel struts, brackets, and bolts. Use four long struts as pillars, and use three shorter ones as cross beams.
   2. Secure the four pillars to the vibration isolation table using brackets and bolts. Connect the gymbal assembly to the middle cross beam using bolts and nuts.
   3. Install a horizontally-oriented load cell in the gymbal assembly; rigidly connect one side of the load cell to the gymbal assembly’s frame, while connecting the other side to the gymbal arm with a wire.
3. Assemble the piezoelectric actuator.
   1. Insert 3 in long, fully-threaded rod through the hole of the piezoelectric stack; put one washer and one nut at each end of the stack; leave about 1/8 in of thread protruding from the end of one nut.
   2. Tighten the nuts at both ends to create a preload in the stack. Connect the long, exposed threads to the gymbal arm using nuts and washers. Thread acorn nut onto the other end of the piezo-actuator and insert disc in the chuck (this acorn nut and disc are used for set-up purposes, not for testing).
   3. Adjust the height of the gymbal assembly so that the acorn nut is in contact with the top of the disc and the gymbal arm is level.
   4. Adjust the position of the gymbal assembly so that the contact point between the acorn nut and disc is about 25 mm away from the rotational center of the disc. Tighten all bolts in the set-up to ensure stability.
4. Set up signal generation, signal amplification, and data acquisition subsystems.
   1. Connect data acquisition system to a lab computer. Connect the output of signal generator to the input of an electrical amplifier. Connect the amplifier output with the input wires of the piezoelectric stack. Connect the amplifier monitors to the data acquisition system.
   2. Connect the load cell to a signal conditioner, and then connect the output of the signal conditioner to the data acquisition system.
5. Additional set-up.
   1. Connect air hose to shop air. Fix the end of the hose to the frame such that its outlet points at the piezo-actuator. Tape the tip of thermocouple to the piezo-actuator. Connect the thermocouple leads to reader; hang the reader on the frame.

2. Pre-test Preparation

1. Calibrate the rotational speed of the motor.
   1. Attach magnet to the rim of the chuck. Place Hall-effect probe close to the chuck. Connect the output of the Hall-effect probe to gaussmeter that is connected to the data acquisition system.
   2. Open the data acquisition software and start data acquisition. Turn on the motor; turn the speed knob of the motor controller to 10 (the lowest rotational speed the motor provides). After the motor rotates for 10 revolutions, turn off the motor. End data acquisition.
   3. Analyze the saved data; the time between two peaks of the output signal from the gaussmeter is the time for the motor to rotate one full revolution.
   4. Turn the knob from 10 to 100 (the highest rotational speed the motor provides) in increments of 10; repeat steps 2.1.2 to 2.1.3.
2. Place load sensor pad between the acorn nut and the disc to measure the normal force at the interface. Finely machine the surface of testing discs using a lathe.
3. Clean the acorn nut and disc to be tested immediately before test.
   1. Put on plastic gloves and face mask.Prepare pieces of lab wipes; fold them into 1 inch squares. Spray ethanol on the tissue squares; gently wipe the surface of the acorn nut and disc with them.
4. Install the clean acorn nut and disc.
   1. Thread the acorn nut onto the piezo-actuator, tighten it with an open-end wrench. Insert the disc in the chuck; adjust the position to make sure the tip of the acorn nut is in contact with the disc surface.
   2. Align the top surface of the disc and the gymbal arm. Tighten the chuck so that the disc is held firmly.
5. Measure the runout of the disc rotation.
   1. Install laser displacement sensor in a fixture, and place the fixture next to the tribometer. Adjust the height and angle of the sensor so that the disc is within the sensor’s range and the laser beam is normal to the disc.
   2. Connect the sensor’s output to the data acquisition system. Start the data acquisition. Turn on the motor and rotate the disc for 10 revolutions; turn off the motor. End data acquisition.

3. Perform Testing

1. Tests with ultrasonic vibrations.
   1. Hang 2 N weight on one hook that connects to the gymbal arm through wire and two pulleys. The weight is used to apply a normal load between the acorn nut and the disc.
   2. Hang another 2 N weight on the other hook that connects to the gymbal arm to provide a horizontal pretension to the load cell.
   3. Set the signal generator to provide a continuous sinusoidal signal with DC offset of 3 V, amplitude of 3 V, and frequency of 22 kHz (the resonance frequency of the piezo-actuator). Note that the 3 V offset is used to prevent tension in the piezo-actuator.
   4. Start data acquisition (reduced friction force). Turn on the amplifier and turn the gain knob to 15, which corresponds to an actual gain of 4.67 (the numbers on the gain knob are arbitrary).
   5. Turn on the motor; set the rotational speed to 6.67 rpm to provide a linear velocity of 20.3 mm/sec. Run the test for 4 hr.
   6. Turn off the motor and amplifier, and then stop the data acquisition. Remove the tested acorn nut and disc from the set-up; Repeat steps 2.3 to 2.5 to install new acorn nut and disc.
   7. Repeat steps 3.1.1 to 3.1.6. In step 3.1.5, set the rotational speed to 13.3 rpm and 28.7 rpm to provide linear velocities of 40.6 mm/sec and 87 mm/sec, respectively; run the tests for 2 and 0.94 hr correspondingly.
2. Tests without ultrasonic vibrations.
   1. Repeat step 3.1.6 to change acorn nuts and discs. Repeat steps 3.1.1 to 3.1.6 with the signal generator and signal amplifier off (the friction measured is intrinsic friction ).

4. Optical Profilometer Measurements

1. Measurement preparation
   1. Clean the discs immediately before measurements using step 2.3. Make eight evenly distributed marks around the rim of the disc. Open the profilometer software.
   2. Raise the lens so that there is sufficient clearance between the lens and sample platform. Level the sample platform. Place a piece of lab wipe on the platform.
   3. Gently place the sample on top of the tissue with one of the eight marks facing the front of the profilometer.
2. Measurement settings.
   1. Choose VSI (Vertical-Scanning Interferometry) as the processing type. Select 5X lens for large field of view and overall shape. Pick 0.55X magnification for a scan area of 1.8 mm by 2.4 mm.
   2. Choose 1X scan speed. Set scan range to -100 m to 100 m. Bring the lens downward toward the sample until there is a blurry image on the screen. Adjust the height of the lens until the image is clear.
   3. Choose 2 as the number of scans to average for each measurement. Click the measurement button.
3. Post-measurement procedures.
   1. Use the vision recipe that defined in the software to correct the raw image for tilt of the whole sample. Open the analysis toolbox in the software.
   2. Obtain the measured roughness values from the “Basic Stats” item. Obtain the measured volume loss of the wear scar within the scan area from the “Volume” item.
   3. Save the images of 1D profiles in x and y directions, the 2D profile, the 3D profile, as well as the table of roughness values. Turn the sample clockwise until the next mark faces the front of the profilometer.
4. Repeat steps 4.2 to 4.3 for the remaining 7 marks.
5. Repeat steps 4.1. to 4.4 on all six discs.

Results

The representative measurements presented here were obtained from the modified tribometer shown in Figure 1. The piezoelectric actuator generates vibrations with amplitude of 2.5 μm at a frequency of 22 kHz. To study the dependence of friction and wear reduction on linear velocity, three different speeds (20.3, 40.6, and 87 mm/sec) were applied to the disc by changing the rotational speed of the motor. For all three groups, the number of disc revolutions and the travel distance of the pin were chose...

Discussion

Experiments were conducted using this protocol to study the effect of linear speed on ultrasonic friction and wear reduction. The measurements show that ultrasonic vibrations effectively reduce friction and wear at three linear speeds. Consistent with prior observations, the amount of friction reduction decreases from 62.2% at 20.3 mm/sec to 29.3% at 87 mm/sec. Wear reduction is negligible with changing linear speed (45.8% to 48.6%).

Material properties such as Young’s modulus and y...

Materials

Name	Company	Catalog Number	Comments
DC Motor	Minarik	SL14
Electrical amplifier	AE Techron	LVC5050
Signal conditioner	Vishay Measurements Group	2310
Signal generator	Agilent	33120A
Piezoelectric stack	EDO corporation	EP200-62
Load cell	Transducer Techniques	MLP-50
Load sensor pad	FlexiForce	A201
Laser meter	Keyence corporation	LK-G32
Hall-effect probe and gaussmeter	Walker Scientific, Inc.	MG-4D
Data acquisition module	Data Physics	Quattro
Data acquisition software	Data Physics	SignalCalc Ace
Thermocouple reader	Omega	HH22
Optical profilometer	Bruker	Contour GT
Profilometer operation software	Bruker	Vision 64