Single-Sided Ultrasonic Welding of CF/Nylon 6 Composite ... · PDF fileglass or carbon fiber)...

WELDING RESEARCH

JANUARY 2018 / WELDING JOURNAL 17-s

Introduction

In recent years, with the develop-ment of manufacturing and processingtechnologies, fiber-reinforced (e.g.,glass or carbon fiber) nylon matrixcomposites have been extensively ap-plied to high technology fields such asaviation/aerospace and automotive in-dustries (Refs. 1–5). A critical step inthe assembly of fiber-reinforced ther-moplastic composite for structural ap-plications is joining. Traditional join-ing methods for metals and ther-mosets, such as mechanical fastening(Ref. 6) and adhesive bonding (Ref. 7),are feasible but not ideal because theycan initiate several irregularities in thestructures. For example, mechanicalfastening may cause fiber breakage(Ref. 6), stress concentration (Ref. 8),delamination (Ref. 9), internal stress

caused by different thermal expansionof the fasteners relative to the com-posites (Ref. 10), and possible galvaniccorrosion (Ref. 11). Adhesive bondingoften requires surface pretreatmentand heat curing (Refs. 7, 11); thestrength of the bonded joints are sen-sitive to the moisture (Ref. 7), temper-ature, and other environmental factors(Ref. 12). Other than mechanical fas-tening and adhesive bonding, a fewstudies in fusion methods (Refs. 10,13–15) demonstrated their potentialfor joining fiber-reinforced thermo-plastic composites. Among all thesepotential methods, ultrasonic weldingis one of the most promising tech-niques in automotive manufacturingbecause it is relatively fast, economi-cal, easily automated, and suitable formass production (Refs. 13, 16–20). Ultrasonic welding is a process

where workpieces are placed between ahorn and anvil. Mechanical vibrationpasses between the horn and anvilthrough the workpieces under a givenhorn force. Heat is generated withinthe workpieces by intermolecular fric-tion at the joint interface, and energydissipation is concentrated at the fay-ing surfaces between the workpieces,forming a weld (Refs. 18–20). The keycontrol parameters in ultrasonic weld-ing are the frequency, amplitude of thevibratory motion, horn pressure, andwelding time (Ref. 21). Although many joint configurationswere used in the ultrasonic welding ofthermoplastics (Refs. 22–24), the lapjoint is the most widely used for auto-motive applications (Ref. 22). In thepractical production, the welding ofthe workpieces may be constrained bypart geometry and designs. Single-sided ultrasonic welding is needed insuch conditions where it is difficult toaccess the backside of the lower work-pieces in conventional ultrasonic weld-ing. Therefore, the difference betweensingle-sided ultrasonic welding andconventional ultrasonic welding is thatthere is no anvil underneath the lowerworkpiece — Fig. 1. In the single-sided ultrasonic weld-ing process, ultrasonic vibration isconducted through the horn, and thevibration wave is transmitted throughthe upper and lower workpieces to theclamp. Whether the absence of ananvil would affect weld formation is amajor concern. To implement single-sided ultrasonic welding, it is neces-sary that the development and under-standing of the single-sided ultrasonicwelding process be obtained. The present study was undertaken to

Single-Sided Ultrasonic Welding of CF/Nylon 6Composite without Energy Directors

An analytical model to explain the weldability of single-sided ultrasonic welded 2.5-mm-thick carbon-fiber-reinforced nylon 6 composite with 30% mass

fiber was formulated

BY L.-Y. CHEN, Q. ZHI, J.-C. LI, Z.-X. LIU, AND P.-C. WANG

ABSTRACT In this study, single-sided ultrasonic welding along with conventional ultrasonicwelding of 2.5-mm-thick carbon-fiber-reinforced nylon 6 composite with 30%mass fiber without energy directors was evaluated. An analytical model toestimate the heat generation in single-sided ultrasonic welding was developed.Testing and modeling results demonstrated that single-sided ultrasonic welded(SSUW) joints had greater weld area and strength than conventional ultrasonicwelds (UWs) under the given process parameters. The differences in weld areaand weld strength were primarily attributed to severe Coulomb friction at the fay-ing surfaces. The bending deformation in single-sided ultrasonic weldinggenerated significant friction heat, resulting in an increase in weld area and weldstrength. This study provided a fundamental understanding of single-sided ultra-sonic welding that would improve the design flexibility of ultrasonic weldedassemblies.

KEYWORDS • Carbon-Fiber-Reinforced Nylon 6 Composite • Weld Formation • Weldability • Single-Sided Ultrasonic Welding • Transient Horn Displacement

Chen (201771).qxp_Layout 1 12/12/17 5:21 PM 7

WELDING RESEARCH

WELDING JOURNAL / JANUARY 2018, VOL. 9718-s

develop a single-sided ultrasonic weld-ing process for joining lapped carbon-fiber-reinforced nylon 6 composite with30% mass fiber. First, we evaluated theeffect of suspension distance on theweldability of single-sided ultrasonicwelding of 2.5-mm- (0.08-in.-) thick car-bon-fiber-reinforced nylon 6 compositewith 30% mass fiber without energy di-rectors. Then, the correlation betweenthe weld formation and transient horndisplacement was analyzed, and the in-fluence of process parameters on the

transient horn displacement was as-sessed. Finally, an analytical model wasproposed and formulated to explain theweld formation mechanism in single-sided ultrasonic welding.

Experimental Procedure

Materials

Commercial pellets of nylon 6 rein-forced with 30 wt-% carbon-fiber com-

posite with a length of 2 mm weresupplied by Poliblend EngineeringPlastics, Italy. The pellets were injec-tion-molded and formed carbon-fiber-reinforced nylon 6 composite with thedimensions of 132 38 2.5 mm (5.2 1.5 0.08 in.). The mechanical prop-erties of the injection-molded work-pieces were measured and the resultsare shown in Table 1.

Ultrasonic Welding Process

The ultrasonic welding process wasperformed using a KZH-2026 multi-function ultrasonic weld (UW) ma-chine (Ref. 25) with a nominal powerof 2.6 kW and a nominal frequency of20 kHz. The output amplitude of thetransducer used in this study was 25m (9.8 10–4 in.), which was fixedduring the welding process. The gainratios of the booster and horn were1:1.2 and 1:1.5, respectively. The dis-placement amplitude of the assembledstack was obtained with a sensor thatwas integrated in the welding ma-chine. Figure 2 shows the welding set-up used in this study. The piezoelectric converter convertsthe electrical signal into mechanical vi-brations. To transfer the ultrasonicwaves to the workpiece, the transduceris connected to the horn that is placedat right angles in contact with the work-pieces to be welded. The support frameof the transducer-booster-horn systemis attached to a pneumatic piston thatprovides vertical movement along withthe static force (i.e., horn pressure) ap-plied through the horn to the work-pieces. The machine was also equippedwith a data acquisition system com-bined with a pressure sensor; a horn-displacement sensor and timer were in-tegrated in the controller of the UW ma-chine, while the horn pressure, weld en-ergy, and displacement of horn wererecorded online in a personal computeras a function of time by the data acqui-sition system. The final horn displace-ment, weld energy, welding time, hornpressure, hold time, and delay time werealso displayed in the control panel dur-ing the ultrasonic welding process. Theworkpieces were clamped with a fixtureas shown in Fig. 2. The machine includes three weldingmodes: energy, time, and horn-displacement modes. The value of theweld energy for energy mode, the

Table 1 — Mechanical Properties of Molded 2.5mm (0.099in.) Thick CarbonFiberReinforced Nylon 6 Composite Coupons with 30% Mass Carbon Fiber

Tensile Strength Elastic Modulus

(MPa) (lb/in.2)) (MPa) (lb/in.2) 89.2 3 (1.3 0.4) 104 7532 412 (1.1 0.6) 106

Fig. 1 — Schematics of ultrasonic welding: A — Conventional; B — single-sided ultra-sonic welding.

Fig. 2 — Schematic of ultrasonic welding of injection-molded carbon-fiber-reinforcednylon 6 composite (dimensions in mm).

A B

Chen (201771).qxp_Layout 1 12/13/17 4:17 PM Page 18

WELDING RESEARCH


welding time for time mode, and thehorn displacement for horn-displace-ment mode were preset to control thewelding process. The workpieces werethen welded using nominal power ofthe machine. To keep the stable con-tact among the horn, upper, and lowerworkpieces at the beginning of ultra-sonic oscillation, the ultrasonic wavegenerator was triggered after the hornwas in contact with the upper work-piece for 2 s (i.e., a trigger time of 2 s).When the weld energy, welding time,or horn displacement reached the pre-set values for the selected weld mode,the ultrasonic wave oscillation wasstopped. Therefore, the weld qualitywas controlled by the preset values ineach selected welding mode.

Single-Sided Ultrasonic Welding

Figure 3 shows a single-sided ultra-sonic welding process. As shown, theworkpiece was placed against anotherworkpiece, with the lower workpieceresting upon a fixture. The horn wasthen placed against the upper work-piece. Ultrasonic vibration was conduct-ed to the horn, and the vibration waveswere transmitted through the upperand lower workpieces to the backup fix-ture that underlined the lower work-piece. A weld was formed at the fayinginterface between the workpieces. To simulate the single-sided ultra-sonic welding process, shims wereused to create a gap of 2 mm betweenthe bottom surface of the lower work-piece and fixture. A lap-shear jointwith the dimensions of 25 38 mm(0.99 1.5 in.) shown in Fig. 3A wasselected in this study. By spacing thedistance between the shims, variousdegrees of single-sided ultrasonicwelding were achieved — Fig. 3B. Not-ed in Figs. 2 and 3, the difference be-tween single-sided ultrasonic weldingand conventional ultrasonic weldingwas that the lower workpiece was sup-ported with one fixed fixture for con-ventional ultrasonic welding (0 mmsuspension distance), while the work-pieces were supported with twoclamps on both ends of the overlap insingle-sided ultrasonic welding. Ultra-sonic welding was performed using theKZH-2026 multifunction ultrasonicwelding machine (Ref. 25) with nomi-nal power and the preset process pa-

rameters (i.e., ultrasonic time, hornpressure, and hold time). When theweld time approached the preset value,the ultrasonic wave oscillation wasstopped and the welded joints werecooled for 5 s. All joints were weldedwith a horn pressure of 0.3 MPa(43.52 lb/in.2) and a welding time of1.3 s using a 7075 aluminum hornwith a diameter of 10 mm. The weldarea and weld strength were used as anindicator of weld quality. The weldarea and weld strength were deter-mined by the average value of threereplicates for each welding condition.

Transient Temperature Measurement

To analyze weld initiation andgrowth during ultrasonic welding, thetemperature evolutions at the loca-tions near the horn-workpiece inter-face and faying surfaces were meas-ured. Figure 4 shows the experimental

setup for temperature measurements.As shown, two small holes with a di-ameter of 0.8 mm (0.032 in.) and adepth of 12.5 mm (0.49 in.), located at0.2 mm (0.008 in.) from the top andbottom surfaces of the upper work-piece, were drilled. Two K-type ther-mocouples were imbedded in twosmall holes and secured with epoxycompound. The temperature evolu-tions at these two locations wererecorded as a function of time by adata acquisition system during ultra-sonic welding.

Quasi-Static Test

Quasi-static tests were performed byloading each specimen to failure in anMTS 810 tensile tester per ASTMD1002-2001 for the joint strength ofthe weld joint. To minimize the bendingstresses inherent in the testing of sin-gle-lap weld specimens, filler platesshown in Fig. 3A were attached to both

Fig. 3 — Schematics of: A — Lap-shear specimen; B — single-sided ultrasonic weld-ing (dimensions in mm).

A

B


WELDING RESEARCH


ends of the specimen using maskingtape to accommodate the sample offset.Load vs. displacement results were ac-quired as the specimens were loaded ata stroke rate of 2 mm/min (0.08 in./min). The joint strength was evaluatedby the peak load. Three replicates wereperformed, and the average peak loadswere reported.

Results and Discussion

Weldability of Single-SidedUltrasonic Welding

For single-sided ultrasonic welding,the workpieces were clamped on both

ends, and there was no anvil on theunderside of the lower workpiece —Fig. 3B. One concern was that ultra-sonic waves would not properly trans-mit and reflect, and form a weld dur-ing ultrasonic welding. To examine the weldability of sin-gle-sided ultrasonic welding of 2.5-mm- (0.099-in.-) thick carbon-fiber -reinforced nylon 6 composite with30% mass fiber without energy direc-tors, experiments were performedwith various suspension distances, re-ferred to in Fig. 3B. Figure 5 presentsthe effect of suspension distance onthe strength of single-sided ultrasonicwelded joints. While the strength ofsingle-sided ultrasonic welded (SSUW)

joints changed little as the suspensiondistance increased from 0 (i.e., con-ventional ultrasonic welding) to 7 mm(0.276 in.) and 15 (0.591) to 19 mm(0.749 in.), it increased significantly asthe suspension distance increasedfrom 7 (0.276) to 15 mm (0.591 in.). To understand why the strengths ofthe SSUW joints were hardly degradedby the absence of an anvil, the weld ar-eas of the SSUW workpieces weremeasured and heat dissipation of theweld controller was recorded. Figure 6presents the effect of suspension dis-tance on the weld area and heat dissi-pation in ultrasonic welding. Asshown, an increase in suspension dis-tance resulted in an increase in weldarea that was closely related to the dis-sipated energy absorbed by the weldedworkpieces. Similar results were ob-served by Bates et al. (Refs. 26, 27).Combining the results shown in Figs. 5and 6 indicated that an increase in dis-sipated energy during ultrasonic weld-ing resulted in an increase in weldarea. These results inferred that weldformation in single-sided ultrasonicwelding likely differed from that ofconventional ultrasonic welding. To understand the increase in weldarea and dissipated energy in single-sided ultrasonic welding, the transienttemperature histories near the horn-workpiece interface and faying sur-faces were measured with the setupshown in Fig. 4, and the results areshown in Fig. 7. As shown, it tookabout 0.8 s for conventional ultrasonic

Fig. 4 — Schematics of the temperature measurements during ultrasonic welding of 2.5-mm- (0.099-in.-) thick lapped carbon-fiber-reinforced nylon 6 composite without energy directors (di-mensions in mm).

Fig. 6 — Effect of suspension distance on the weld area and dissipated energy inultrasonic welding of 2.5-mm- (0.099-in.-) thick carbon-fiber-reinforced nylon 6composite without energy directors.

Fig. 5 — Effect of suspension distance on the strength ofultrasonic welded 2.5-mm- (0.099-in.-) thick carbon-fiber-reinforced nylon 6 composite without energy directors.


WELDING RESEARCH


welding and 0.5 s for single-sided ul-trasonic welding to reach the meltingpoint of nylon 6 (i.e., 240°C/464°F)near the faying surfaces, respectively.The higher peak temperature andheating rate observed in single-sidedultrasonic welding suggested thatmore dissipated energy was absorbedby the SSUW joints (Refs. 28, 29),which was consistent with the resultsfound in Fig. 6.

Transient Horn Displacementin Ultrasonic Welding

In Fig. 7, the suspension distanceinfluenced the temperature evolutionsthat may correlate with weld initiationand growth. To understand how thesuspension distance would influencethe weld growth mechanism in ultra-sonic welding, the transient horn dis-placements were measured with thedata acquisition system connected tothe UW machine, and the results arepresented in Fig. 8. As shown, thehorn moved downwards under a givenhorn pressure during ultrasonic weld-ing as the result of heat generation atthe faying surfaces, which softenedand melted the workpieces. Therefore,the weld growth evolutions were likelyreflected by the variations of transienthorn movement. Careful examinationof the results shown in Fig. 8 indicatedthat the transient horn displacementduring ultrasonic welding could be di-vided into four phases (Refs. 30, 31).

Phase I was the solidCoulomb frictionstage ranging from 0to ~0.3 s; Phase II was

the unsteady penetration stage rang-ing from 0.3 to ~0.8 s; Phase III wasthe steady penetration stage rangingfrom 0.8 to ~1.3 s; and Phase IV wasthe cooling stage starting from 1.3 s.In addition to these four phases, at thebeginning of the single-sided ultrason-ic welding process, there was an initialhorn displacement, identified as Phase0 (0 to 0.3 s), which increased with anincrease in suspension distance. Com-pared to the conventional ultrasonicwelding process, as the suspension dis-tances were above 11 mm (0.433 in.)in single-sided ultrasonic welding, therate of increase in the penetrate phaseincreased significantly. These resultssuggested that the weld not only initi-ated early but also grew rapidly in sin-gle-sided ultrasonic welding, whichwere consistent with the findings ob-served in Fig. 7.

Weld Formation in UltrasonicWelding

As the results described, the weldinitiated early and grew rapidly as thesuspension distance increased in single-sided ultrasonic welding. Becausethere was no anvil underneath the lowerworkpiece, the workpieces were underbending under an applied horn pressureduring single-sided ultrasonic welding.This affected the contacts at the fayingsurfaces of the workpieces, and thus in-fluenced the weld formation (Ref. 29).

To understand how the suspension dis-tance affected the weld formation, thefracture surfaces of the quasi-static test-ed ultrasonic welded specimens were examined. Figure 9A–E show the weld areasfor the welds made with the suspen-sion distances of 0, 7, 11, 15, and 19mm (0, 0.276, 0.433, 0.591, and 0.749in.), respectively. As shown, while asingle weld was fabricated in conven-tional ultrasonic welding, multiple iso-lated welds were formed for a suspen-sion distance of 7 mm (0.276 in.), re-ferring to Fig. 9A and B. These resultsdemonstrated that the welds madewith a suspension distance of 7 mm(0.276 in. a 10-mm (0.394-in.) diam-eter horn) had a similar welding mech-anism to that of conventional ultra-sonic welding. Because the upper andlower workpieces were constrained un-der the horn or near the clamps, theresistance to vibrations rose for theworkpieces adjacent to the clamps.

The ultrasonic waves likely trans-mitted through the clamp-to-work-piece interface and led to the weldedworkpieces vibrating, and consequent-ly produced the major welds along theclamps. As the suspension distance in-creased to 11 mm (0.433 in.), quitedifferent findings were observed inFig. 9C. While the minor welds formedalong the clamps, the major weldswere formed at the faying surfaces un-derneath the periphery of the horn-to-workpiece interface. These results sug-gested that there likely existed anoth-er welding mechanism other than theconventional ultrasonic welding mech-

Fig. 7 — Transient temperature-time histories of the nearhorn-workpiece interface and faying surface of ultrasonicwelding (0 mm suspension distance) and single-sided ul-trasonic welding (19 mm (0.749 in.) suspension distance)of 2.5-mm- (0.099-in.-) thick carbon-fiber-reinforcednylon 6 composite.

Fig. 8 — Effect of suspension distance on the horn displacementin ultrasonic welding of 2.5-mm- (0.099-in.-) thick carbon-fiber-reinforced nylon 6 composite.


WELDING RESEARCH


anism. Referring to Fig. 9C, the work-pieces began to bend downward undera given horn pressure when the sus-pension distance of 11 mm (0.433 in.)was slightly larger than the horn diam-eter (i.e., 10 mm (0.394 in.)). Theworkpieces deformed easier in the

width directionthan in the lengthdirection (i.e.,longer axis of theworkpiece), andthe bending de-formation causedan additional rel-ative movementon the faying sur-faces along thetangent direction

of the deformed workpieces during ul-trasonic oscillations. The heat generated due to Coulombfriction resulting from relative move-ment of the workpieces could be com-bined with the heat produced due to in-termolecular friction to melt the mate-

rials at the faying surfaces. It is notedthat the bending deformation in thewidth direction resulted in an asymmet-rical contact at the faying surfaces (la-beled as B in Fig. 9C), which likely led toan intimate contact (at site B) betweenthe workpieces. As discussed earlier, theconstrained site between the upper andlower workpieces had large resistance tovibrations, and the power dissipated inultrasonic welding was related to thepressure at the contact regions (Refs.21, 30, 31). Hence, the heat was gener-ated and formed the weld pools at siteB. Meanwhile, two minor welds (labeledA in Fig. 9C) were resulted from theheat produced due to intermolecularfriction at the faying surfaces near twoclamps.

Fig. 9 — Effect of suspension distance on the fracture sur-faces of ultrasonic welded 2.5-mm- (0.099-in.-) thick lapcarbon-fiber-reinforced nylon 6 composite: A — 0 mm (i.e.,ultrasonic welding); B — 7 mm (0.276 in.); C — 11 mm (0.433in.); D — 15 mm (0.591 in.); and E — 19 mm (0.749 in.) (dimen-sions in mm).

A

C

E

B

D


WELDING RESEARCH


With an increase in suspension dis-tance from 11 (0.433) to 15 mm (0.591in.) and 19 mm (0.749 in.), more bend-ing deformation was produced under agiven horn pressure, and consequentlythe contact areas between the horn andupper workpiece gradually evolved toform a weld — Fig. 9D and E. The weldinitiated at the faying surfaces near theperiphery the of horn-to-workpiece in-terface (i.e., labeled A in Fig. 9D and E)and grew toward the clamps rather thanforming near the clamps. These resultssuggested that the heat due to Coulombfriction was produced at the faying sur-faces as the suspension distance in-creased. Furthermore, Coulomb frictionwas more severe in the width directionthan in the suspension direction due totheir difference in bending deforma-tion, which resulted in more frictionheat in the width direction. Additional-ly, the weld formed at the intersectionof the clamps and the periphery of thehorn-to-workpiece interface as shownin Fig. 9D and E. These results inferredthat the change of suspension distanceaffected the contacts between the horn-to-workpiece and workpiece-to-work-piece, and consequently the heat gener-ation at the faying surfaces.

Modeling of Single-SidedUltrasonic Welding

As the results described, the work-pieces deformed during squeeze cycleprior to ultrasonic oscillations, and the

deformed workpieces sprung backonce the horn was retrieved. It was as-sumed that a relative motion devel-oped at the faying surfaces along bothwidth and length directions. The tan-gential and normal motions betweenthe workpieces led to Coulomb frictionand intermolecular friction, respec-tively, which resulted in heat genera-tion at the faying surfaces during ul-trasonic welding. To understand theheat generation, an analytical modelshown in Fig. 10 was derived. For sim-plification, the workpieces betweentwo clamps were assumed deformedindependently in directions of couponwidth and length. The deformed re-gions in width and suspension direc-tions of the workpieces were repre-sented as ADW and ADS, respectively —Fig. 10B. A ring-like superimposed de-formed region (plaid region in Fig. 10)was formed between the horn andclamps, in which the materials wereunder severe deformation and had arelative tangent movement at the fay-ing surfaces during ultrasonic welding. Referring to Figs. 9 and 10, as thesuspension distances increased from11 to 19 mm (0.433 to 0.749 in.), theworkpieces had more compliance inwidth direction than in the length di-rection and consequently deformedeasier in the width direction than inthe suspension direction. The normalmotion and tangential motion at su-perimposed regions ASSUW (dark plaidregion ABCA in Fig. 10B) generated

heat due to intermolecular frictionand Coulomb friction of the work-pieces, which melted the workpieces atthe faying surfaces. The weld area,ASSUW, can be estimated from the areadifference between the sectoral regionOABC and triangular region OAC inFig. 10B as follows:

where L is the suspension distance andD0 is the horn diameter. By substitut-ing the horn diameter (i.e., 10 mm(0.394 in.)) and suspension distances(i.e., 11 (0.433), 15 (0.591), and 19mm (0.749 in.)) into Equation 1, itclearly shows that an increase in sus-pension distance resulted in an in-crease in weld area (ASSUW), whichagreed with the experimental resultsobserved in Fig. 9C–E. Once the weld formation in single-sided ultrasonic welding was analyzed,the next question was how the heatgeneration at the faying surfaces wasinfluenced by the changes in suspen-sion distance. Careful examination ofthe experimental setup indicated thatthe changes in suspension distancewould result in complex horn pressuredistribution on the workpieces, and afinite element analysis would be desir-able to understand the correlation be-tween the suspension distance and

Assuw = L2

4�arc cos

D0L

��

��

��

��

�D04

L2 �D02 (1)

Fig. 10 — Schematics on the effect of suspension distance on the weld formation inultrasonic welding of 2.5-mm- (0.099-in.-) thick carbon-fiber composite: A — 11 mm(0.433 in.); B — 19 mm (0.749 in.) (dimensions in mm).

Fig. 11 — Sketch of force and ampli-tude analysis during squeeze cycle insingle-sided ultrasonic welding.

BA


WELDING RESEARCH


heat generation. However, due to thelack of the coefficients of friction be-tween the objects (e.g., the work-pieces-to-clamps), a simplified analy-sis was adopted in this study. Examination of the results in Fig. 9Cand D indicated that the weld forma-tion near the periphery of the horn-to-workpiece interface (i.e., regions A)likely came from the heat due toCoulomb friction resulting from thebending deformation of the workpiecesunder a horn pressure. Since Coulombfriction heat in single-sided ultrasonicwelding of polymeric composite is likelysimilar to that of ultrasonic welding ofmetallic workpieces, the heat genera-tion model (Refs. 32–34) for metallicworkpieces was adopted in this study. Per the model of Elangovan (Ref.32), Ding (Ref. 33), and De Vries (Ref.34), the heat generated, Wmetal, in ultra-sonic welding of metallic material wasprimarily from the deformation ofworkpieces and friction at the fayingsurfaces of the workpieces, which canbe expressed as follows (Ref. 34):

where a0 is the horn amplitude; f is thehorn frequency; P0 is the horn pressure;and s, , and Aw are the yield strength,coefficient of static friction, and weldarea of metallic material, respectively. Referring to 10B, Coulomb frictionin this study was from the tangentmovement of the workpieces at thefaying surfaces during ultrasonic oscil-lation, and the resultant heat formed aweld in the dark plaid region ABCA.Because the workpieces had a fixedwidth (38 mm), the horn pressure andamplitude in single-sided ultrasonicwelding can be estimated using thefour-point bending beam shown inFig. 11. As shown, the vertical compo-nent (P0n) of the horn pressure P0, andthe normal (a0n) component and tan-gential component (a0t) of the hornamplitude (a0) can be expressed, re-spectively, as follows:

where is the angle of rotation of theworkpiece under an applied horn pres-sure in single-sided ultrasonic welding,which can be estimated per classicalmechanics of material:

where L is the suspension distance, D0

is the horn diameter, and El is therigidity of the welded workpieces. Re-placed P0 and a0 in Equation 2 with P0n

in Equation 3 and a0t in Equation 4, re-spectively, the power dissipated,Wcoulomb, in the sectoral region ABCA ofthe SSUW workpieces (Fig. 10B) togenerate Coulomb frictional heat canbe expressed as follows:

Because the heat generated due tointermolecular friction from the nor-mal motion of the workpieces was likethat of conventional ultrasonic weld-ing of polymer, the power dissipatedto produce intermolecular friction, Wintermolecular, is related to the frequencyof ultrasonic oscillation, normal ampli-tude (a0n) of the horn, and loss modu-lus of the carbon-fiber nylon compos-ite shown below (Ref. 30):

where E” is the loss modulus of thecarbon-fiber composite. The combina-tion of Equations 1, 6, 7, and 8 re-vealed that an increase in suspensiondistance resulted in an increase in an-gle of rotation of the workpiece ()and Assuw. Therefore, as the power dis-sipated for generation of the Coulombfriction heat increased (i.e., Equation7), the power dissipated for producingthe intermolecular friction heat de-creased (i.e., Equation 8). The totalpower dissipated, WSSUW, in single-sided ultrasonic welding of workpiecescan be obtained as

The results in Fig. 6 indicated thatan increase in suspension distance re-

sulted in an increase in dissipatedpower (WSSUW) in SSUW workpieces.However, based on Equations 7–9, anincrease in suspension distance result-ed in an increase in Wcoulomb (Equation8) but a decrease in Wintermolecular (Equa-tion 7). Thus, the increase in WSSUW

likely resulted from the increase inWcoulomb that overrode the decrease inWintermolecular as the suspension distanceincreased, which led to additional heatat the faying surfaces.

Summary

In this study, the weldability of single-sided ultrasonic welding of 2.5-mm- (0.099-in.-) thick carbon-fiber-reinforced nylon 6 composite with 30%mass fiber without energy directorswas studied. Modeling and experimen-tal results demonstrated that the sin-gle-sided ultrasonic welded (SSUW)joints possessed greater heat genera-tion, weld area, and weld strength thanconventional ultrasonic welded jointsfor the given process variables. The useof this process can increase the designand manufacturing flexibility of ultra-sonic welded assemblies. However, thesingle-sided ultrasonic welding processdeveloped in this study had some no-table limitations in practical applica-tion. The bending deformation of theworkpieces in single-sided ultrasonicwelding induced coulomb friction thatwould likely produce warpage andresidual stresses in the welded work-pieces. The warpage and residual stress-es may result in dimensional issues andweakened mechanical strengths of thewelded workpieces. Therefore, morestudies, such as part assembly and fa-tigue strength of the SSUW workpieces,are required before the implementationof the process for practical application.

Conclusions

1) For the given process variables,the SSUW 2.5-mm- (0.099-in.-) thickcarbon-fiber-reinforced nylon 6 com-posite had greater heat generation, weldarea, and weld strength than the con-ventional ultrasonic welded joints. 2) The bending deformation insqueeze cycle in single-sided ultrasonicwelding of 2.5-mm-thick carbon-fiber-reinforced nylon 6 composite resultedin severe contacts at the faying surfaces

Wintermolecular =2 fa02E" cos (8)

Wcoulomb = 4fa0 sin Assuw

s / 2( )2 = P0 cos / Assuw( ) / 2( )2deformation power( )+ 1

2μP0 sin2

friction power( ) (7)

P0n = P0 cos (3)

a0t = a0 sin (4)

a0n = a0 cos (5)

=P0L L2 D0

2( )16El

(6)

Wmetal = 4a0f

Aws / 2( )2

= P0 / Aw( ) / 2( )2+μP0 (2)

WSSUW = Wintermolecular +Wcoulomb (9)

Chen (201771).qxp_Layout 1 12/13/17 11:02 AM Page 24

WELDING RESEARCH


and likely produced significant Coulombfriction under ultrasonic oscillations,and consequently generated significantfriction heat that was increased with anincrease in suspension distance (i.e.,distance between the clamping fixture)in single-sided ultrasonic welding. 3) The welds formed in single-sidedultrasonic welding of 2.5-mm-thick car-bon-fiber-reinforced nylon 6 compositeresulted from the combination ofCoulomb friction heat from relativemovement of the workpieces and inter-molecular friction heat at the fayingsurfaces.

The authors gratefully acknowledgethe financial and technical support pro-vided by GM Global Research and De-velopment to carry out the presentwork.

1. Fuchs, E. R. H., Field, F. R., Roth, R.,and Kirchain, R. E. 2008. Strategic materialsselection in the automobile body: Economicopportunities for polymer composite de-sign. Composites Science and Technology68(9): 1989–2002. 2. Strong, A. B. 1993. High Performanceand Engineering Thermoplastic Composites.Lancaster, Penn.: Brent Strong TechnomicPub.Co. 3. Zah, R., Hischier, R., Leao, A. L., andBraun, I. 2007. Curaua fibers in the automo-bile industry — A sustainability assess-ment. Journal of Cleaner Production 15(11,12): 1032–1040. 4. Duflou, J. R., De Moor, J., Verpoest, I.,and Dewulf, W. 2009. Environmental im-pact analysis of composite use in car manu-facturing. CIRP Annals – ManufacturingTechnology 58(1): 9–12. 5. Orsato, R. J., and Wells, P. 2007.U-turn: The rise and demise of the automo-bile industry. Journal of Cleaner Production15(11, 12): 994–1006. 6. Vinson, J. R. 2004. Mechanical fasten-ing of polymer composites. Polymer Engi-neering and Science 29(19): 1332–1339. 7. Vinson, J. R. 1989. Adhesive bondingof polymer composites. Polymer Engineeringand Science 29(19): 1325–1331. 8. Strong, A. B. 1993. High Performanceand Engineering Thermoplastic Composites.

Lancaster, Penn.: Brent Strong TechnomicPub. Co. 9. Todd, S. M. 1990. Joining thermoplas-tic composites [M]. Proceedings of the 22nd

International SAMPE Technical Conference,vol. 22, pp. 383–392. 10. Ageorges, C., Ye, L., and Hou, M.2001. Advances in fusion bonding tech-niques for joining thermoplastic matrixcomposite: A review. Composites: Part A32(32): 839–857. 11. Barnes, T. A., and Pashby, I. R. 2000.Joining techniques for aluminum space-frames used in automobiles: Part II — Ad-hesive bonding and mechanical fasteners.Journal of Materials Processing Technology99(1–3): 72–79. 12. Davies, P., Courty, C., Xanthopoulos,N., and Mathieu, H.-J. 1991. Surface treat-ments for adhesive bonding of carbon fiber-poly (etherether ketone) composites. Jour-nal of Materials Science Letters 10(6):335–338. 13. Grewell, D., and Benatar, A. 2007.Welding of plastics: Fundamentals and newdevelopments. International Polymer Process-ing 22(1): 43–60. 14. Hou, M., Yang, M., Beehag, A., Mai,Y.-W., and Ye, L. 1999. Resistance welding ofcarbon fiber reinforced thermoplastic com-posite using alternative heating element.Composite Structures 47(1–4): 667–672. 15. Ahmed, T. J., Stavrov, D., Bersee, H.E. N., and Beukers, A. 2006. Induction weld-ing of thermoplastic composites — Anoverview. Composites: Part A 37(10):1638–1651. 16. Levy, A., Poitou, A., Le Corre, S., andSoccard, E. 2008. Ultrasonic welding ofthermoplastic composites, modeling of theprocess. International Journal of MaterialForming 1(Supplement 1): 887–890. 17. Villegas, I. F., and Bersee, H. E. N.2010. Ultrasonic welding of advanced ther-moplastic composites: An investigation onenergy-directing surfaces. Advances in Poly-mer Technology 29(2): 112–121. 18. Levy, A., Le Corre, S., and Villegas, I. F. 2014. Modeling of the heating phenom-ena in ultrasonic welding of thermoplasticcomposites with flat energy directors. Jour-nal of Materials Processing Technology 214(7):1361–1371. 19. Qiu, J., Zhang, G., and Wu, Y. 2009.Proposal of ultrasonic welding techniqueand weld performances applied to polymers.Polymer Engineering and Science 49(9):1755–1759. 20. Chang, J., Zheng, C., and Ni, Q.-Q.2006. The ultrasonic wave propagation incomposite material and its characteristicevaluation. Composite Structures 75(1–4):

451–456. 21. Glowska, A. W., and Pietras, A. 2012.Influence of the welding parameters on thestructure and mechanical properties of vi-bration welded joints of dissimilar grades ofnylons. Archives of Civil and Mechanical Engi-neering 12(2): 198–204. 22. Rani, M. R., Prakasan, K., andRudramoorthy, R. 2005. Designing jointsfor ultrasonic welding of plastics. WeldingJournal 84(9): 50-s to 54-s. 23. Devine, J. 2001. Ultrasonic plasticwelding basics. Welding Journal 80(1): 29-sto 33-s. 24. Stokes, V. K. 2000. Assessment ofgeometries for determining strengths ofthermoplastic vibration welds. Journal ofMaterials Science 35(10): 2393–2403. 25. Weihai Kaizheng Ultrasonic Tech-nologies Co. Ltd., China. 26. Bates, P. J., MacDonald, J. J., Wang,C. Y., Mah, J., and Liang, H. 2003. Vibrationwelding nylon 66 — Part I: Experimentalstudy. Journal of Thermoplastic CompositeMaterials 16(2): 101–119. 27. Bates, P. J., Dyck, C., and Osti, M.2004. Vibration welding of nylon 6 to nylon66. Polymer Engineering and Science 44(4):760–771. 28. Sahin, O. S., Koellhoffer, S., Gillespie,J., Advani, S., and Bogetti, T. 2014. Thermalmodeling during continuous ultrasonicwelding. Turkish Journal of Engineering & Sci-ences 38(1): 79–96. 29. Khmelev, V. N., Slivin, A. N., andAbramov, A. D. 2007. Model of process andcalculation of energy for a heat generationof a welded joint at ultrasonic welding poly-meric thermoplastic materials. 8th SiberianRussian Workshop and Tutorial on ElectronDevices and Materials. 30. Potente, H. 1984. Ultrasonic welding— Principles and theory. Materials & Design5(5): 228–234. 31. Chung, Y.-M., and Kamal, M. R.2008. Morphology of PA-6 vibration weldedjoints and its effect on weld strength. Poly-mer Engineering and Science 48(2): 240–248. 32. Elangovan, S., Semeer, S., andPrakasan, K. 2009. Temperature and stressdistribution in ultrasonic metal welding —An FEA-based study. Journal of MaterialsProcessing Technology 209(3): 1143–1150. 33. Ding, Y., Kim, J., and Tong, P. 2006.Numerical analysis of ultrasonic wire bond-ing, effects of bonding parameters on con-tact pressure and frictional energy. Mechan-ics of Materials 38(1, 2): 11–24. 34. De Vries, E. 2004. Mechanics andMechanism of Ultrasonic Metal Welding[M], Dissertation. The Ohio State University.

Acknowledgments

References

LONG-YANG CHEN, QIAN ZHI, JIAN-CUN LI, AND ZHONG-XIA LIU ([email protected]) are with the Key Lab of Materials Physics,School of Physics and Engineering, Zhengzhou University, Zhengzhou, Henan, China. PEI-CHUNG is with the Manufacturing SystemsResearch Lab, General Motors Research ans Development Center, Warren Mich.


Single-Sided Ultrasonic Welding of CF/Nylon 6 Composite ... · PDF fileglass or carbon fiber)...

Documents

Transcript of Single-Sided Ultrasonic Welding of CF/Nylon 6 Composite ... · PDF fileglass or carbon fiber)...