Product Performance
Damping characteristics of unreinforced,glass and carbon
?ber reinforced nylon6/6spur gears
S.Senthilvelan,R.Gnanamoorthy*
Department of Mechanical Engineering,Indian Institute of Technology Madras,Chennai600036,India
Received3August2005;received in revised form12September2005
Abstract
Polymer based gears replace metal gears in many light duty power and/or motion transmission applications due to their noiseless operation even under unlubricated conditions.Visco elastic behaviour of the polymer,which is mainly responsible for the sound absorption,is altered by the addition of short?bers.Addition of high modulus?bers to the base polymer matrix reduces the damping characteristics of the composite.Material damping affects the hysteresis heating as well as the ability to absorb vibration during service.Dynamic mechanical analysis carried out on unreinforced Nylon6/6,20%short glass and20%carbon?ber reinforced Nylon6/6gear materials indicates the reduction of damping factor due to the incorporation of?bers.Injection molded spur gears made of unreinforced and reinforced materials were tested for durability in a power absorption type gear test rig.Surface temperature of the test gears and noise generated near the gear mesh region were continuously measured and monitored using a personal computer based data acquisition system.Test results indicate that the reinforced gears generate more gear mesh noise than unreinforced gears.
q2005Elsevier Ltd.All rights reserved.
Keywords:Polymer composite;Short?ber;Gear;Hysteresis;Damping;Noise
1.Introduction
Noise generation in a gear transmission system depends upon many factors such as gear quality, contact ratio of the gear pair,gear material resilience and operating conditions.In a light duty application,if noiseless operation is the main objective,then polymeric materials that have good damping charac-teristics are preferred.Addition of high strength and modulus?bers decreases the viscous component of a composite and enhances the elastic component. Reduction of the viscous component affects the material damping capabilities.Polymer composite gears are used for many light and medium duty power transmission as well as motion transmission appli-cations.Many research works were carried out to analyze the in?uence of short?ber reinforcement on the fatigue life,wear characteristics and ef?ciency of the polymer composite gears[1–4].However,the damping characteristics of the polymer composite gears have received little investigation.Tsukamoto et al.[5] studied the acoustics of polyamide gears.Addition of plasticizer to the Nylon material reduced the noise generation during service.In order to reduce the noise and vibration,plastic material has been?tted in between the toothed peripheral rim and inside the boss made of steel[6].Calabrese et al.[7]replaced steel timing gears with plastic gears in a diesel engine and compared the noise emission.Luscher and Houser[8] investigated the geometry and transmission errors
of *Corresponding author.Tel.:C914422574691;fax:C9144
22570545.
E-mail address:gmoorthy@iitm.ac.in(R.Gnanamoorthy).
injection-molded gears with different gating con?gur-ations.Uematsu[9]investigated the effect of gear tooth pro?le error on the angular velocity variation of a transmission.Adams and Maheri[10]reported the in?uence of?ber orientation and stacking sequence on the damping behaviour of anisotropic beams and plates made of?ber-reinforced polymers.Pritz[11]charac-terized the damping resistance of a material using magnitude and width of the loss peak factor.
Damping in the gear tooth also affects the amount of hysteresis heat generation during service.Many works have been reported on the heat generation and its in?uence on gear performance[12–16].Repeated gear tooth de?ection causes internal friction within the material as well as contact(surface)friction due to meshing of the gear teeth[12].Yousef and Burns[13] developed a test rig for fatigue testing of thermoplastic gear teeth.The temperature and bending fatigue strength of polyacetal and polycarbonate gears were measured.The molecular motion within the viscoelas-tic polymeric material dissipates the internal energy in the form of heat due to gear tooth bending,sliding, shearing and impact between mating steel gears.Since most of the polymeric gear materials are poor thermal conductors,generated heat will not be easily dissipated, resulting in reduction of gear strength[14].Since the properties of polymer-based materials are sensitive to temperature,rise in surface temperature deteriorates the mechanical and wear characteristics of the gear.Kof?et al.[15]developed a model to predict heat generation in thermoplastic gears.Hooke et al.[16]examined the wear behaviour of polymeric gears and relate the gear tooth surface temperature with gear wear.
This paper describes the in?uence of short?ber reinforcement on damping behavior of thermoplastic composite gears.Glass and carbon?ber reinforced injection molded polyamide base gears were used for the studies.
2.Dynamic mechanical analysis
The dynamical mechanical analysis was carried out to investigate the effects of temperature and frequency on viscoelastic properties of polymer and polymer https://www.doczj.com/doc/e913216373.html,mercially available injection mold-able unreinforced,20%glass?ber reinforced and20% carbon?ber reinforced Nylon6/6granules were used for the studies.Specimens with dimensions of60! 10!4mm were molded in a standard two plate single pin pointed gate mold in the laboratory.Dynamic mechanical analysis was carried out using test equipment(NETZCH)according to ISO6721-5[17].Specimens were subjected to three-point bending with a span length of40mm.An oscillating force was applied (maximum3N)to give constant amplitude of de?ection of120m m.Measurements were conducted over the temperature range of305to448K with a heating rate of 2.0K/min,and under a constant frequency of1.0Hz.By measuring the time lag of the displacement to the applied force,the damping factor of the material was determined.
3.Test gear details and processing conditions
Unreinforced,20%glass?ber reinforced and20% carbon?ber reinforced Nylon6/6granules were preheated for4h at353K to lower the moisture content before injection molding.A three-plate mould with three symmetrical pin pointed gates was used for molding unreinforced and reinforced gears.The three-plate mould has the advantage of separating the runner from the gear in the mold and it allows an axial injection path,which results in a more concentric gear. Test gears with2mm module,208pressure angle,17 teeth,and6mm face width were made in the laboratory using an injection-molding machine at a molding pressure of125MPa and melt temperature of543K. Injection molded test gears were mated against the hobbed stainless steel(SS316)gears in a power absorption type gear test rig as well as a static gear tooth de?ection test rig.Molded gears were inspected using a fully automatic CNC controlled gear-measuring center (Klingelnberg P40).
4.Static gear tooth de?ection test rig
Since stress and strain are not in phase for a viscoelastic material,the stress–strain curve forms a loop.Determination of the hysteresis loss of a polymer material is a most common method for determining the magnitude of internal friction generated during cyclic loading[14,18,19].A test rig was designed and developed to measure the gear tooth de?ection under actual gear meshing conditions.Measured single tooth de?ection under loading and unloading quanti?es the hysteresis loss and the results were correlated with the net surface temperature of test gears measured during the gear performance tests.
Fig.1a shows the photograph of the test unit.In this unit,the applied load is not a point load as in the case of conventional single tooth loading test rig.Load is applied through the meshing gear similar to the actual gear transmitting system.The rig consists of two shafts; test gear shaft and mating gear shaft,which are
S.Senthilvelan,R.Gnanamoorthy/Polymer Testing25(2006)56–6257
mounted on a block at the standard center distance of the gear pair.The mating gear shaft is mounted in a block such that it can rotate about its axis with a pair of supporting bearings.At one end of the mating gear shaft,a standard stainless steel gear is mounted rigidly through which the load is transferred to the test gear.A loading arm is ?xed rigidly at the other end of the mating gear shaft.A pan is ?xed at the one end of loading arm for holding the dead weights,and an adjustable counter weight is placed at the other end so that the weight of the empty pan and lever can be counter balanced.The test gear is mounted rigidly on the test shaft.
When dead weights are added to the loading arm,the mating gear shaft carrying the steel gear tends to rotate.Since the steel gear is meshing with the test
gear,a known amount of torque acts on the test gear.As the test gear is rigidly ?xed,the applied torque de?ects the gear tooth.The linear de?ection of the test gear tooth is measured using a dial gauge with 1m m accuracy.A tooth is removed from the test and mating steel gears to facilitate the position of dial gauge plunger.The removed adjacent tooth in both the test and mating gears does not contribute to the load sharing in the actual meshing condition.Molded unreinforced,glass ?ber reinforced and carbon ?ber reinforced gears were considered for the studies.Corresponding dead weights were added on the loading pan to apply torques of 0.5,1.2and 1.8N m.The chosen loading range is well within the elastic range of test gear materials.The spring-loaded plunger of a precision dial gauge was positioned at the pitch region of the test gear so that the linear de?ection of the gear tooth at the pitch line is measured.Fig.1b shows the close up view of meshed test and mating gear teeth under test conditions.
5.Power absorption type gear test rig
Fig.2a shows schematically the developed power absorption type gear test rig used for evaluating the gear performance.In this rig,the test gear is driven using a direct current (DC)motor.The test gear mates with an identical standard steel gear,which is connected to the DC generator.The required test torque is introduced by loading the rheostat connected to the generator.Some other sources of sound generation during tests apart from gear mesh excitation are the DC motor,DC generator,coupling and bearings.Fundamental mesh frequencies of any of the above-mentioned system were not considered and,hence,no ?lter was employed.The microphone of a precision sound level meter (B &K make)was kept at a distance of about 50mm from the gear mesh region to measure the sound pressure at a frequency of 1Hz.Molded test gears were tested at a constant rotational speed of 1000rpm.Gears were tested at 1.5and 2.0N m torque until failure or 5million cycles,whichever was earlier.Non-contact infrared temperature sensors were mounted at a distance of about 5mm from the land surface of the test gear (Fig.2b).The net surface temperature of test gears and sound level were monitored continuously using a personal computer based data acquisition system.Detail damage mechanisms of the test gears and condition monitoring techniques adoped are discussed elsewhere [12,20]
.
Fig.1.Developed static gear tooth de?ection test rig (a)photograph of test rig and (b)close up view of the test gear tooth in mesh.
S.Senthilvelan,R.Gnanamoorthy /Polymer Testing 25(2006)56–62
58
6.Results and discussion
6.1.Damping characteristics of test gear materials Dynamic mechanical analysis conducted on test gear materials indicates a rise in the storage modulus and a drop in the damping factor due to the incorporation of short reinforced ?bers to the Nylon 6/6matrix (Fig.3a).Due to the superior modulus of carbon ?ber compared with the glass ?ber,carbon ?ber reinforced Nylon 6/6material shows a higher storage modulus compared with the glass ?ber reinforced Nylon 6/6composite.At the glass transition temperature of the test materials,storage moduli of unreinforced,glass ?ber reinforced and carbon ?ber reinforced Nylon 6/6materials are 1336,5525and 10,944MPa,respectively.Fig.3b shows the drop in damping factor with the addition of ?bers.At the glass transition temperature,tan d values of unreinforced,glass ?ber reinforced and carbon ?ber reinforced Nylon 6/6materials are 0.1321,0.0756and 0.0547,respectively.6.2.Hysteresis of test gears
Due to the material hysteresis,de?ection at a particular load does not show the same magnitude
Test frequency: 1 Hz
02500
50007500100001250015000
(a)(b)303
328
353378403428453
Temperature (K)
303
328
353378403428453
Temperature (K)
S t o r a g e m o d u l u s (M P a )-
20 % Carbon fiber reinforced Nylon 6/6
20 % Glass fiber reinforced Nylon 6/6
Un reinforced Nylon 6/6
0.000
0.0300.0600.0900.1200.150
D a m p i n g f a c
t o r (t a n δ)
Fig.3.Dynamic mechanical analysis of test gear materials (a)storage modulus and (b)damping factor.
1. Motor
2. Generator
3. Test gear
4. Steel gear
5. Torque sensor
6. Temperature sensor
7. Vibration Pickup
8. Microphone
9. Speed sensor 10. Counter
11. Coupling
12. Bearing
Microphone
Temperature sensor
(a)
(b)
Fig.2.(a)Schematic of power absorption type gear test rig and (b)close view showing temperature sensors and sound level meter.
S.Senthilvelan,R.Gnanamoorthy /Polymer Testing 25(2006)56–6259
during loading and unloading.In all the investigated test gears,a difference between the loading and unloading path is observed.Fig.4shows the hysteresis loop obtained for unreinforced,glass ?ber reinforced and carbon ?ber reinforced Nylon 6/6gears.The area enclosed by the loading and unloading path quanti?es the hysteresis loss of test gears.Hysteresis loss of the unreinforced Nylon 6/6gear tooth is taken as unity.Hysteresis loss of glass ?ber reinforced Nylon 6/6gear tooth is 0.67times that of the unreinforced Nylon 6/6gear tooth,whereas energy lost by the carbon ?ber reinforced Nylon 6/6gear tooth is 0.60times that of the unreinforced Nylon 6/6gear tooth.Hence,less heat generation during service is expected for reinforced gears.
6.3.Surface temperature of test gears
Meshing of load bearing members experience heating due to both internal friction and contact friction.The coef?cient of friction of polymeric materials is extremely low and less heat will be generated due to the contact friction.Therefore,surface temperature rise during the gear performance (or heat generated)is mostly due to gear tooth repeated bending and the associated material hysteresis heating.The amount of heat generation in gear tests depends upon three major parameters;gear material damping,gear rotational speed,and transmitting torque.In the current investi-gations,the operating conditions (speed and torque)are the same for all the gear tests investigated and,hence,the only major in?uencing parameter of heat generation is gear material hysteresis damping.A considerable amount of generated heat is also dissipated to the atmosphere by convection,since the test gears are running at 1000rpm.Measured surface temperature of the test gear is the algebraic sum of generated and
dissipated heat.Fig.5(a)and (b)shows the measured temperature of unreinforced and reinforced gears tested at 1.5and 2N m torque.With increase in torque,both the unreinforced and reinforced test gears showed increased surface temperature due to the larger tooth de?ection.At 1.5N m torque,all test gears show a rise in temperature up to 50,000cycles beyond which a steady state is reached,whereas at 2N m loading all the test gears show a sudden rise in temperature up to 20,000cycles,beyond which a steady state is reached.At both the test torque levels,reinforced gears show a lower surface temperature compared with unreinforced gears,with carbon ?ber reinforced gears being lower than glass ?ber reinforced gears.6.4.Gear noise
Gears during service experience dynamic loads and the magnitude of dynamic loads depends on the quality of the gears,running speed,transmitting torque and gear material [21].The noise and vibration level of the gear unit depend on the dynamic loads.Fig.6(a)and (b)shows the measured sound pressure near the gear mesh region during performance testing for a period of 0.2million cycles at different loading conditions.At both 1.5and 2.0N m test torque levels,reinforced
gears
Fig.5.Measured surface temperature of unreinforced and reinforced gears tested at (a)1.5N m and (b)2N m.
Applied torque (Nm)
D e f l e c t i o n o f s i n g l e t o o t h (μm )
Fig. 4.In?uence of reinforcement on gear tooth stiffness and hysteresis losses.
S.Senthilvelan,R.Gnanamoorthy /Polymer Testing 25(2006)56–62
60
generated more noise than unreinforced gears,with carbon?ber reinforced gear generating more than glass ?ber reinforced gears.The lower damping factor of the carbon?ber reinforced Nylon6/6material compared with the glass?ber reinforced Nylon6/6material is responsible for this increased noise generation.
Gear quality in?uences the dynamic loading and thereby the noise generation.Detailed gear tooth inspection carried out on the molded gears is described in detail in Ref.[22].Addition of short?bers to Nylon6/ 6material causes anistopic shrinkage,which consider-ably increases the lead and pitch error.The asymmetric nature of the?bers restricts the shrinkage of polymer matrix in the direction of the?ber orientation,and the highest shrinkage occurs in the direction transverse to orientation.Fig.7shows the worst(maximum)pitch deviation measured in unreinforced and reinforced gears.The difference in tooth deviation observed between glass and carbon?ber reinforced Nylon6/6 gear is due to the variation in mold shrinkage observed, which in turn is due to the difference in?ber density and ?ber matrix compatibility.
Higher dynamic load is generated in low accuracy gears.An approximate method developed by Niemann is used to calculate the dynamic load.The dynamic load (Ft dyn)in a gear tooth which is responsible for the noise and vibration level is given by[21]
F tdyn Z K0K v bv(1) K0Z
F t
b
f!26f(2) where F tdyn is the additional tangential force on the tooth caused by dynamic loads(N),K0is the factor taking into account the amount of load in relation to the effective tooth errors(N/cm),K v is the factor taking care of the in?uence of the circumferential velocity(s/m),b is the face width of tooth(cm),v is the circumferential velocity (m/s),F t is the tangential force on the tooth(N)and f is the largest existing tooth error(m m).
The worst tooth pitch variation was taken as the largest existing tooth error,as this deviation was the maximum of the gear tooth parameters considered. Using Eq.(2),K0is computed for the applied torque of 1.5N m.The values of K0are1290,2798and2538N/ cm for unreinforced,glass?ber reinforced and carbon ?ber reinforced Nylon6/6gears,respectively.Corre-sponding to the magnitude of K0and v,the value of K v is obtained from the standard[21].Fig.8shows the magnitude of dynamic force based on the measured gear tooth error and gear velocity tested.Higher dynamic loads are generated in the reinforced gears compared with the unreinforced gears.This clearly indicates that the noise generated in gears is also due to the lower accuracy of reinforced gears.
7.Conclusions
Incorporation of glass and carbon?bers to the Nylon 6/6material decreases the damping capacity of the composite.Reduction of the viscous component due to the reinforcement causes a reduction of
material 0
30
60
90
120
Unreinforced
Nylon 6/6
Carbon reinforced
Nylon 6/6
Gear materials
Glass reinforced
Nylon 6/6
W
o
r
s
t
p
i
t
c
h
e
r
r
o
r
(
μ
m
)
Fig.7.Measured worst(maximum)pitch error of unreinforced and reinforced test
gears.
Fig.6.Sound level measured during gear performance tests at(a)
1.5N m and(b)2N m.
S.Senthilvelan,R.Gnanamoorthy/Polymer Testing25(2006)56–6261
resilience.Damping characteristics of the gear material in?uences the noise and heat generation during service.Unreinforced Nylon 6/6gears exhibited more hysteresis loss compared to glass and carbon ?ber reinforced Nylon 6/6gears,with glass ?ber reinforced gears having more hysteresis loss than carbon ?ber reinforced gears.
Fiber reinforcement reduces the internal heat generation capacity in the polymer matrix during cyclic loading and gears reinforced with high modulus ?bers show a lower heat generation during service.However,gears reinforced with high modulus ?bers generated higher noise compared with unreinforced gears.Acknowledgements
Authors thank Dr S.K.Malhotra and Dr R.Velmurugan of Composite Technology Centre for the various help during the course of the project work.References
[1]M.Kurokawa,Y.Uchiyama,S.Nagai,Performance of plastic
gear made of carbon ?ber reinforced poly-ether-ether-ketone,Tribology International 32(1999)491–497.
[2]D.Walton, A.B.Croppe, D.J.Weale,P.Klein Meuleman,
The ef?ciency and friction of plastic cylindrical gears:part 1—In?uence of materials,Journal of Engineering Tribology.Proceedings of the IMechE Part J 216(2002)75–92.[3]N.A.Wright,S.N.Kukureka,Wear testing and measurement
techniques for polymer composite gear,Wear 250(2001)1567–1578.
[4]S.Senthilvelan,R.Gnanamoorthy,Wear characteristics of
injection molded un?lled and glass ?lled Nylon 6spur gears,Journal of Engineering Tribology Proceedings of the IMechE—Part J 218(2004)495–502.
[5]N.Tsukamoto,H.Maruyama,T.Ikuta,Development of low
noise gears (silencing effect and operating characteristics of polyamide elastomer gears),JSME International Journal 34(1991)245–250.
[6]S.Takanashi,H.Shiraishi,Characteristics of gear ?tted with
plastic material for shock absorbing Proceedings of International Symposium on Gearing and Power Transmissions,Tokyo 1981pp.479–484.
[7]A.Calabrese,P.Davoli,S.Quaranta,Noise reduction with
plastic gears;a case history for diesel engines Proceedings of Second International Conference on Gearbox Noise,Vibration,and Diagnostics,London 1995pp.133–140.
[8]A.Luscher, D.Houser, C.Snow,An Investigation of the
geometry and transmission error of injection molded gears,Journal of Injection Molding Technology 4(2000)177–190.[9]S.Uematsu,Effect of variation of angular velocity in gear
rolling process on pro?le error,Precision Engineering 26(2002)425–429.
[10]R.D.Adams,M.R.Maheri,Damping in advanced polymer
matrix composites,Journal of Alloys and Compounds 355(2003)126–130.
[11]T.Pritz,Loss factor peak of viscoelastic material:magnitude to
width relation,Journal of Sound and Vibration 246(2001)265–280.[12]S.Senthilvelan,R.Gnanamoorthy,Damage mechanisms in
injection molded unreinforced,glass and carbon reinforced nylon 6/6spur gears,Applied Composite Materials 11(2004)377–397.
[13]S.S.Yousef,D.J.Burns,Techniques for assessing the running
temperature and fatigue strength of thermoplastic gears,Mechanism and Machine Theory 8(1973)175–185.
[14]V.R.Dominick,V.R.Donald,G.R.Marlene,Plastics Design
Handbook,Kluwer Academic Publishers,Dordrecht,2000.[15]D.Kof?,R.Gauvin,H.Yelle,Heat generation in thermoplastic
spur gears,Journal of Mechanism,Transmission and Auto-mation in Design 107(1985)31–37.
[16]C.J.Hooke,K.Mao,D.Walton,A.R.Breeds,S.N.Kukureka,
Measurement and prediction of the surface temperature in polymer gears and its relationship to gear wear,Transactions of ASME,Journal of Tribology 115(1993)119–124.
[17]ISO 6721-5:1996,Plastics—Determination of dynamic mechanical
properties—Part 5:Flexural vibration—Non-resonance method.[18]W.Van Paepegem,K.De Geyter,P.Vanhooymissen,J.
Degrieck,Effect of friction on the hysteresis loops from three-point bending fatigue tests of ?ber-reinforced https://www.doczj.com/doc/e913216373.html,posite Structures,in press (Corrected Proof,Available online 21December 2004).
[19]A.M.Kucherskii,Hysteresis losses in carbon-black-?lled
rubbers under small and large elongations.Polymer Testing,24(2005)733–738.
[20]S.Senthilvelan,R.Gnanamoorthy,Condition monitoring and
failure analysis of composite gear Proceedings of 11th National Conference of Machines and Mechanisms,New Delhi 2003pp.278–282.
[21]G.M.Maitra,Handbook of Gear Design,Tata McGraw-Hill
Pub.Ltd.,New Delhi,2003.
[22]S.Senthilvelan,Development and performance of injection
molded polymer composite gears,PhD Thesis (under revision,2005).
20406080100Unreinforced Nylon 6/6 gear Carbonreinforced
Nylon 6/6 gear
Gear type
Glass reinforced Nylon 6/6 gear
D y n a m i c f o r c e (N )
1.5 Nm
2.0 Nm
Fig.8.Dynamic force generated due to gear tooth error and operating speed in test gears.
S.Senthilvelan,R.Gnanamoorthy /Polymer Testing 25(2006)56–62
62