Shoulder desain .pdf

Journal of Manufacturing Processes 20 (2015) 1523

Contents lists available at ScienceDirect

Journal of Manufacturing Processes

j ourna l h o me page: www.elsev ier .com/ locate /manpro

Technical Paper

Shoulder design developments for FSW lap jointspolyme

Shayan E reirINEGI, Institute as, 40

a r t i c l

Article history:Received 18 MReceived in re16 SeptemberAccepted 29 SAvailable onlin

Keywords:Friction stir wPolymer weldiTool designStationary shoAxial forcePolypropylenePolystyrene (PS)

velopms torticu

deveaterilyse qymerity anrce m

factu

1. Introduction

Friction promising jdevelopmeFSW was into weld usiate benetsinto the tecpart, surpripropagationincreased th

Given thdesign strucnew challenmost attraclosophy andtechnique w[2], the met

CorresponE-mail add

(T. Ramos), pta(P.M.G.P. More

such as polymers, metallic materials, copper [3,4] and even joining

http://dx.doi.o1526-6125/ stir welding (FSW) technique is one of the mostoining methods, which accounts for the considerablent efforts the technique has gone in the last decade.troduced to join lightweight alloys that are difcultng conventional techniques. Although some immedi-

of the technique were planned in advance and builthnique, such as the high-quality nishing of the weldedsing side benets such an improved resistance to crack

relative to the base material were found that furthere interest in the development of the tool.e recent increase in industrial demand for lightweighttures [1], novel joining methods are required to tackleges, such as multi-material joining. FSW is one of thetive methods in this regard, due to its solid-state phi-

the ability for full automation. Even though the FSWas originally developed for joining aluminium alloys

hod is presently being studied to weld other materials

ding author. Tel.: +351 913853305.resses: [email protected] (S. Eslami), [email protected]@inegi.up.pt (P.J. Tavares), [email protected]).

dissimilar materials [5].Among all the advantages of the FSW technique, the following

stand-out clearly due to their potential impact in industry:

- Being a solid-state welding process; the generated heat for thisprocess is about 70% to 90% the base materials melting point [6];

- It is applicable to components of a large range of thicknesses withaccurate reproducibility [7,8];

- Doesnt require post-welding processing;- Due to the low amount of heat generated, components with high

mechanical properties, low distortion and residual stresses areobtained;

- It is an inherently environmental-friendly process because noller material, toxic fumes or shielding gases are employed orgenerated;

- Traditional welding defects such as hot cracking and porosity arenot an issue [9,10].

The FSW process itself is based on the heat generated by frictionbetween the FSW tool and the base material surface [11]. Conven-tional FSW tools consist of a rotating pin attached to a shoulder,which penetrates the parts to be welded and traverse along theweld line, as demonstrated in Fig. 1. The generated heat leads to

rg/10.1016/j.jmapro.2015.09.0132015 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved.rs

slami , Tiago Ramos, Paulo J. Tavares, P.M.G.P. Mo of Science and Innovation in Mechanical and Industrial Engineering, R. Dr. Roberto Fri

e i n f o

arch 2015vised form

2015eptember 2015e 26 October 2015

elding (FSW)ng

ulder

(PP)

a b s t r a c t

This paper describes the process of depolymers. Previous FSW research seethat leads to consistent results, in papolymers. This study is focused on thestir welding of dissimilar polymeric mtries have been tested in order to anastationary shoulders made out of poltool improved the welds surface qualto improve the stability in the axial foa conventional FSW tool.

2015 The Society of Manu of dissimilar

a0 4200-465 Porto, Portugal

ing a stationary FSW tool for welding thin plates of dissimilar be still far from the type of systematic engineering approachlar for polymeric materials and especially for lap joining oflopment of welding tools, aiming at sound and robust frictionals in lap joint conguration. Different materials and geome-uality of the welds and appearance. It has been veried that

materials give the best result. The welds produced with thisd strength signicantly. The use of the proposed tool showedagnitude during the welding procedure in comparison with

ring Engineers. Published by Elsevier Ltd. All rights reserved.

16 S. Eslami et al. / Journal of Manufacturing Processes 20 (2015) 1523

extreme stoccurs and

During tthe weld linow and blocated on tand avoids smooth surtwo areas: isame as thethe side whrotational drial is transfby the shou

The conand compoindustrial ain strengthpolymers athe techniqtageous forjoining is cu

Weldinglowing clas

(1) Mechanfriction

(2) Additioresistan

For all thenough heathe nearly-mthe nal sta

Authorslute solid-spolymers drange. In vimelting poisequently, twith a relawill be enou

The maingenerated tmaterial. Thorder to incFor FSW of ing shoulde

stir the material together. Therefore, the shoulder has an essentialrole in this process and is one of the parameters that plays a consid-erable effect on the weld strength, as well as the welding surface.It is worth mentioning that a good welding surface is normally a

e inde stualityonal hardals caeveloncerchniqidereouggineted

oductry a

in wrbonding], a cW oties aechnld setrenglly ated hwhice areers, this m

to lecene, deer wbe tal in

quag higped clow nicalgeneed piance

worerenrene

the Fig. 1. FSW schematic representation.

irring, reaching a state in which plastic deformationis appropriate for welding [12].he welding process, the rotating probe positioned insidee causes the material to become soft and enables it to

e stirred. On the other hand, the rotating shoulder ishe surface of the plates, which generates frictional heatthe material ash leakage out of the seam, providing aface. As illustrated in Fig. 1, the weld bead is divided inn the advancing side the tool rotational direction is the

welding direction; the retreating side corresponds toere the tool welding direction is opposite to the toolirection. In order to weld the two plates, the soft mate-erred with the probe, along with a forging force causedlder.sumption of lightweight materials, such as polymerssites has been growing dramatically in a diversity ofnd engineering applications, due to the improvement-to-weight ratio of those materials [13]. Consequently,nd composite materials joining are the new challengesue needs to address. Just as FSW proved its advan-

metallic materials, a breakthrough in FSW polymersrrently sought after.

of polymeric materials can be categorized into the fol-sications [14]:

ical movement to generate heat: vibration welding; welding; ultrasonic welding;nal heating technique: hot gas; hot plate; implant andce welding.

e above welding methods, the rst step is to generatet in order to reach the plastic deformation stage, thenolten material will bond together under pressure and

ge will be to let the material cool down [15,16].

positivSom

face qutraditipoint, materinew dthat cothis teis cons

Althing enattempthe prgeomedefects20% caby bontion [8and FSproperother tthe weweld sgenerageneraareas, in thospolymbead. Tleadinguntil rhot shoshoulding promaterisurfaceweldindeveloto the mechaon the threadappear

Theof diffPolystystrates such as Strand claim that polymer FSW is not an abso-tate process [16]. Due to different molecular weights,o not have a particular melting point, but a meltingew of this fact, shorter polymer chains can reach theirnt during welding, whereas longer chains may not. Con-he soft material will be the mixture of molten materialtively small amount of solid material. However, theregh molten material to let the material ow easily.

difculty for FSW polymers is the lack of frictional heathrough contact between the rotational tool and the baseis applied friction should generate the adequate heat inrease the material temperature near its melting point.Aluminium alloys, this task is implemented by a rotat-r touching the surface, which generates enough heat to

polymeric mformation o

2. Experim

Friction CNC machimethod wiwhich is shminium ana more sopinstrumentequipmentication of a high joint quality.dies realized [17] that it is difcult to obtain a good sur-

and high mechanical properties of the welded part withFSW tools in polymers welding. Due to the low meltingness and thermal conductivity of polymers [18], thesen reach their melting point quickly, thereby demandingpment which are still not available, particularly thosen development of tools. Since tool plays a critical role inue, development of sufcient FSW tools for polymersd a topic that needs further investigations.h FSW of thermoplastic composites is a challeng-ering eld, only a restricted number of researcherswelding polymers using FSW [19]. Tensile strength ofed welds is very low and mostly affected by the pinnd welding speed, which brings about the formation ofelds on Polypropylene (PP) composites reinforced with

bre (CF). PP and its composites are typically joined or welding methods [20,21]. In a previous investiga-omparison between the common welding techniquesf PP has studied, clearly showing that the mechanicalchieved by FSW are higher than the ones obtained withiques. Moreover, they established that the roughness ofam is higher than that of the parent material and theth is lower than half the one of the base material. Theccepted explanation for this behaviour is based on theeat distribution at the weld seam and its surrounding

h leads to a non-uniform crystallization rate of materialas. In another study [22], it was realized that for FSW ofhe traditional tools push the soft material out of the weldaterial loss is responsible for poor bonding formation,

ow tensile strength and poor mechanical properties. Uptly, the most common tool used with polymers FSW isveloped by Nelson et al. [22]. This tool consists of a staticith a heater and a thermocouple inside, and a rotat-o stir the almost molten material. This tool traps theside the weld bead, promoting the formation of a goodlity. More recently, new tools have been developed forh density polyethylene by Kiss and Czigny [8]. The toolonsists of a hot shoe with a threaded pin. However, due

thermal conductivity of polymers, it was detected that properties of the obtained welds are highly dependentrated heat. In another study [23], a heated shoulder withn was used for butt joining ABS plastics, and a good weld

was obtained with this tool.k presented in this manuscript focuses on the effectt tool designs on lap joints of dissimilar polymers:

and Polypropylene. Furthermore, this study demon-advantages of using a stationary shoulder for weldingaterials, and suggests a possible path to prevent the

f defects and achieve high quality welds.

ental procedure

stir welding experiments were implemented on a 3-axisne. The welds were produced using a position controlth a sensorized clamping system for load acquisition,own in Fig. 2. Polymers behave differently than alu-

d tend to buckle easier under pressure, which demandshisticated clamping system. The designed xture ised with four load cells connected to data acquisition. The data is acquired and manipulated with a dedicated

S. Eslami et al. / Journal of Manufacturing Processes 20 (2015) 1523 17

F on the

Table 1Material speci

Material a)

PolypropylePolystyrene

LabViewTM

observationRotation

parametersFrom theseimentationsgeometry p

Dissimilfor the lap jteristics of t

Since thFSW tool fothe literatuin an iteratusing the foto 70 mm/m800 rpm to

Differencourse of tha conventioAfterwardscarried out.rating beari

als hiumwithhe sthe s

Fig. 3. Some oshoulders withig. 2. Fixing system: (a) isometric view, (b) front view (c) real image of the xture

cations.

Dimensions (mm3) Tensile strength (MP

ne (PP) 100 80 1.5 28 (PS) 100 80 2.6 50

code developed in our laboratory that enables the and registration of the applied force during welding.al speed, traverse speed and the tool geometry are

which can most prominently affect the weld strength. parameters, and based on a set of preliminary exper-

materialumintested

In tinside for FSW of polymers, it was veried that the toollays a signicant role on the joint overall quality.ar polymer materials and thicknesses have been usedoints produced in this work. Table 1 shows the charac-he materials which have been used.is study is focused on the development of a suitabler polymers, and since no similar study can be found inre, different tool designs and dimensions were analysedive path, Fig. 3. The new tool designs were developedllowing parameters ranges: welding speed 10 mm/minin, pin length 2.8 mm to 3.6 mm, and rotational speed

3000 rpm, depending on the tool being used.t probes and shoulders have been developed in thee current work. The initial tests were performed usingnal rotating shoulder design with different diameters., a rotational probe with a stationary shoulder study was

In order to avoid the shoulder rotation, a tool incorpo-ngs was designed. Stationary shoulders with different

venting thestationary sinside the sthat lead shto avoid thipin. Ejectedthe retreatifurther stud

The streperformed a 2 mm/minwith a prop

3. Results

The mainew FSW tosurface qua

f the FSW tools developed: (a) from left to right, wood, teon, aluminium and polycarbona different diameters. CNC table and (d) detail of the load cells conguration.

Melting range (C) Density (kg/m3)

163 910180280 1040

ave been developed and analysed. Polycarbonate, teon,, wood and brass stationary shoulders have all been

different degrees of success, Fig. 3.tationary shoulder solution, ball and thrust bearingshoulder let the rotating probe rotates with spindle pre- shoulder to spin. One of the main challenges of using ahoulder is to prevent the injection of the soft materialhoulder and bearing, and it is one of the main reasonsoulders to fail, in particular during long runs. In orders problem, different sleeves have been used around the

material from the weld seam can lead to root defects inng side of the weld, which is not acceptable and needsy.ngth of the welds was determined in tensile testingin an Instrom ElectroPulsTM machine, Fig. 4(b), using

loading rate. Specimens were cut out of the weld seamrietary tool developed for this study, Fig. 4(a) and (c).

and discussion

n objective of the presented study is the development ofols capable of achieving quality welds with acceptablelity. The tools developed in the course of the present

te stationary shoulders, (b) brass stationary shoulder and (c) rotating


ess te

study havediameter, aformed in ofor the diffestirring, ten

One of ta good surfmaterial, thThe amounthe used tooetration anthe materiastationary sprevious inerated heatFor this resand tested observe thesurface qua

3.1. Conven

The mostogether wiating heat dface of the heat can beon the pareor, heat ovemodify the to the tool material fro

An initiasquared pinand a dama

Since thshoulders wtheir largernot possiblefaces were welds the shand stick ar

It was tnot a good

a) Welded parts with 3 mm conventional FSW tool and (b) FSW tool withuared probe and 8 mm rotating shoulder.

ts a very rough surface and ash and root defects occurredhe weld line.

ationary shoulder

ce, the results obtained using conventional shoulders weretisfactory, a stationary shoulder design was developed tote the weld bead surface quality. Using a stationary shoul-e results were better than those obtained with the previousg tool design. A schematic design of the stationary shoulderped is presented in Fig. 7.Fig. 4. (a) Schematic of overlap area of the specimen, (b) tensile str

all been tested for a number of probe length, probend rotational and traverse speeds. Peel tests were per-rder to evaluate the weld bead materials congurationrent probe geometries. For those that led to an improvedsile tests and analysis were carried out.he most important characteristics of a quality weld isace quality. The more similar the weld is to its parente more it retains the parent materials characteristics.t of heat generated by friction is directly dependent onl. Geometric features of the pin and shoulder, tool pen-

d diameters are the main factors which directly affectl ow and heat generation during welding [24]. Using ahoulder, all the frictional heat is generated by the pin. Investigations, in order to compensate for the lack of gen-, extra heating hot shoe systems have been employed.earch, different pins and shoulders have been createdwithout any extra heating on the plates in order to

effect of the different tools, particularly on the resultinglity.

tional shoulder

t popular tool used in FSW consists on a probe rotatingth the shoulder. The shoulder has the main role of gener-ue to the friction between the shoulder and the top sur-workpiece. The direct effect of the generated frictional

Fig. 5. (3 mm sq

presenalong t

3.2. St

Sinnot saevaluader, throtatindevelo observed on the weld width. Nevertheless, dependingnt material, insufcient heat can prevent proper joiningrow can lead to partial melting. Partial melting willmaterial properties as well as bond the melted materialsurface. In order to achieve strong welds, the ejectedm the weld seam has to be kept at a minimum.l 8 mm diameter shoulder has been tested with a 3 mm. As can be seen in Fig. 5, the material was not mixedged surface occurred for all the welding parameters.e lack of heat often leads to insufcient bonding, largerith 10, 15, 20 mm diameters, have been tested. Due to

diameter, additional heat was generated, but it was still to obtain appealing weld surfaces. Although good sur-obtained at the initial length of some welds, for longeroulder literally melted all the material causing it to slip

ound the probe and shoulder, as it can be seen in Fig. 6.herefore concluded that a rotating shoulder design isoption for welding polymers. The produced weld bead

Howevethe heat, f

Fig. 6. (a) We(b) PP materiast and (c) specimen ready to be tested.r, without the rotating shoulder, the pin generates allrom the rotational movement. Most of the previous

lded part with 3 mm squared probe and 20 mm rotating shoulder andl stuck around the probe and shoulder surface.


Fig. 7. Stationary tool schematic representation.

studies in this topic, used extra heating from within the shoulder,in order to compensate for the inherent lack of heat [16].

A stationary shoulder made with wood and a ball bearing wastested rst, and the initial results were promising, as it can be seenin Fig. 8(a). However, some defects were still present. Wood pre-sented deformation under pressure, which caused defects on theretreating side of the weld bead. Nevertheless, it was possible toavoid the ash defect, characteristic of a stationary shoulder design.The stationary shoulder pushes the material down and does not letthe soft material go outside the bead, letting it cool down under theapplied pressure, and preventing the appearance of such defects.

In someobserved wrial was foupressure, sothe bearingit was realiimproveme

In ordershoulder, dthe shouldeshort shoulas smooth ader was conto be weldebeen notice

Using a surface quaarea aroundof the pin tefor long weshoulder cafurther defesimilar welunder differ

Finally, shown in Fi

ool reing an

t of ing uck o

s funr thes a tack oy devrder to avoid material loss, brass and copper sleeves withtion free capability were tested. Copper, alone as a sleeve,xcessively and deforms the Teon around the pin. Brass cane same problem, but in some cases, the brass sleeve rotatede pin under pressure and caused the shoulder to melt around. It was concluded that the most effective solution is to have aleeve inside a copper sleeve, Fig. 11. It gives one more degreedom and even though the brass sleeve rotates, it does sothe copper sleeve and will not damage the shoulder. This

Fig. 8. We welds, a lack of the PP material in the weld line washich caused root defects. Usually, the missing mate-nd inside the shoulder and bearing, cf. Fig. 9. Underft materials go upwards around the pin, then inside

and damage the bearing. Despite promising results,zed that a stationary shoulder design needs additionalnts in order to achieve defect free welds.

to overcome the difculties found with the woodenifferent materials were tested to evaluate the effect ofr material on the weld surface. Using an aluminium

der, the surface appearance was relatively good, but nots required, Fig. 8(b). A long aluminium stationary shoul-sequently tested with the ability to heat up the platesd in advance, although no obvious improvement hasd.Teon shoulder allowed the improvement of the weldlity signicantly. However, in some circumstances, the

the probe suffered large deformation as a consequencemperature. This shoulder deformability usually occurs

lds with a low transverse speed, Fig. 8(c). Any deformedn cause material loss inside the weld seam leading tocts. The polycarbonate shoulder, Fig. 8(d), presented ad quality as the Teon, but with lower deformabilityent welding circumstances.a fully brass stationary shoulder has been tested asg. 10, but due to its high thermal conductivity, a large

Fig. 9. Tball bear

amounof heatfrom la

Thirials foremaining a lthereb

In olubricaheats ehave thwith ththe pinbrass sof freeinside lded part with: (a) wooden stationary shoulder; (b) aluminium stationary shoulder; (c) tlated issues. (a) set of damaged tool parts; (b) PP material inside thed (c) PP material inside the shoulder, around the probe.

heat was accumulated inside the brass volume insteadp the bead, and consequently the weld nugget sufferedf sufcient heat generation.ctional procedure enabled the choice of the best mate-

shoulder in Polycarbonate or Teon although there stillendency for the materials to move up the bearing, caus-f material in the weld bead. An improved design waseloped to address this issue.eon stationary shoulder; (d) polycarbonate stationary shoulder.


Fig. 10. (a) Weld seam created by brass stationary shoulder and (b) Stationary shoulder fully made by brass.

method can easily increase the life of the shoulder and eliminatesdefects such as lack of material in the weld bead.

3.3. Probe

With the absence of a rotating shoulder to assist in the genera-tion of heat, the probe plays an even more important role in orderto produce heat and stir the soft material together. To weld poly-meric materials, probes should have grooves or threads to allow the

softened material to ow simply with excessive amount of turbu-lence. Without grooves the material sticks on the advancing side ofthe weld and will not stir sufciently. Threads have the same effectas grooves, with improved friction followed by additional heat. Thedirection of the thread should be opposite of the spindle rotationdirection, otherwise the threads will move the soft material outsidethe weld bead.

Three different probe geometries were tested: cylindrical withtwo at surfaces, triangular and squared probes have all been

Fig. 11. Newlysleeves, (d) we developed FSW tool with stationary shoulders (a) nal design drawing, (b) actual tool lded part with teon stationary shoulder and 6 mm probe and (e) overall view of the devduring welding, (c) teon stationary shoulder with brass and coppereloped FSW tool.


Fig. 12. Set of different probes tested.

Fig. 13. Differences between non-threaded probe (a) and threaded probe and (b)after the peel test.

investigated in this study. A set of different tools tested, geometryand diameter, is presented in Fig. 12.

These probe geometries were selected as a result of theirincreased angle of attack and consequently their ability to gener-ate more heat. The resulting difference between non-threaded andthreaded pin can be seen in Fig. 13. While in Fig. 13(a) both translu-cent and black materials are easily perceptible in the weld line, inFig. 13(b) they are not.

Probe diameter and length are other effective factors on the weldquality that should be considered. If the difference between thick-ness of the plates and probe length is out of range, the second plate,which is PS, will not be joined properly due to lack of penetra-tion, and defects may occur in the weld bead. Likewise, the use of ashort probe may result in insufcient heat generation and impropermaterial stirring. The investigated pins have a range of 2 to 3.6 mmin length and 3 to 7 mm in diameter.

3.4. Axial force

Because axial force plays a critical role in FSW process, the quan-tication of the applied force for different types of tool design canhelp understand their behaviour during the welding operation. Itwas veried that the axial force is a function of the tool being used.Due to the large area of the stationary shoulder, when the shoul-der contacts the plates the applied force increases signicantly, butwhen the tool advances the force decreases and stabilizes evenusing a position control system. This behaviour can be explainedby the effect of temperature on the material hardness. The axialforce for Teon stationary shoulder with 3 and 7 mm triangularprobes and different welding speeds is shown in Fig. 14. Further-more, the use of threaded pins will reduce the axial force during

Fig. 14. Stationary shoulder axial force during welding.


Table 2Loads before failure for tested specimens*.

Specimen ID Shoulder penetration [mm] Pin length [mm] Traversing speed [mm/min] Rotational speed [rpm] Maximum Load [N]

R3 0.1 2.05 50 1700 83.3S3 N/A 3.40 30

* All the tests were performed with 3 mm probe diameter and overlapped area of 600 mm2.

welding. Using a conventional rotating tool an unstable loading isveried when a position control system is used. Axial force using aconventional FSW tool was found to be dened in three main zonesor sections, which can be distinguished by different force peaks andslopes. While the rst zone is characterized by a low force unstablestate, the second presents a high slope followed by the maximumaxial force, or peak, decreasing rapidly to the third section where itremains relatively stable and constant.

3.5. Tensile

Lap sheashoulder joTable 2 shousing a conpared to thTwo specimthe maximresults demusing the stwhen comperal weldinstrongest wcations mthe Table 2achieved wmal conducthe retreatifailed fromobtained usbase materi

4. Conclus

A new dpresented idesign, Tabuated, and developed. tries and pato ash defdesign the f

Table 3Overall view o

Rotatory shodiameter (m

8 10

12 15

20

The statioface quali

A Teon pbest optioto alumin

Pins mus Using sta

during thing sest rhoulee ofder ttionaasedetreaheat der tthe r

wled

s wo1759gia 4d MC

nces

heri pertiep://dxligan ld J 19ghan

usinp://dx

ar lding p://dxaeili tion

of p://dxasawa

frictio0;210s ZCT

stirp://dxs Z, terialsme.20kerso

root42-11 strength

r tests were performed for both stationary and rotatingints to quantify the tool effect on the weld strength.ws the tensile strength results of specimens weldedventional tool with a 3 mm squared probe (R3) com-ose welded using the Teon stationary shoulder (S3).ens were cut for each set of welding parameters, andum loads in the Table 2 are their mean values. Theonstrate that the tensile strength of specimens obtainedationary shoulder tool can have more than 40% strengthared to those achieved with a rotating shoulder. Sev-g parameters have been tested in order to achieve theeld using conventional tool. The strongest weld speci-ade by rotating shoulder can be seen in the rst row of. The obtained results using stationary shoulders wereithout parameters optimization. Due to the low ther-tivity of polymers and insufcient heat generation onng side of the welds, during the tests all the specimen

the retreating side. The tensile strength of the jointsing a stationary shoulder is approximately 50% of PPal (28 MPa tensile strength, Table 1).

ion

esign for polymeric material FSW was developed andn the present manuscript. The effect of different toolle 3, shoulders and pins, on the weld quality was eval-an optimized tool design for welding polymers wasUsing a conventional FSW tool with different geome-rameters did not enable production of sound welds dueects around the weld bead. Using a stationary shoulderollowing conclusions can be drawn:

f the tested shoulders and probes.

ulderm)

Stationary shouldermaterial

Probe geometries

Polycarbonate CylindricalTeon Threaded cylindrical

(M6, M3 screws)

rotat The b

ary sdegrin or

Funcincre

The rand

In order,

Ackno

ThiTME/1TipoloESF an

Refere

[1] Bagprohtt

[2] ColWe

[3] Dehperhtt

[4] Kumwehtt

[5] Esmtigainghtt

[6] Hiring201

[7] Kistionhtt

[8] Kismapp.

[9] Dictains01Wood TriangularAluminium (short andlong)

Triangular with groove

Brass SquareCopper Conical with two at

surfaces

[10] Barcellonring in f2006;177

[11] Mendes ogy and 2014;58:1800 117.0

nary shoulder enabled stronger welds with good sur-ty.olymeric material stationary shoulder proved to be then, resulting in superior surface quality when comparedium, brass and wood shoulders.t have grooves in order to properly stir the material.tionary shoulders, the applied force is kept constante operation, the same behaviour cannot be found usinghoulders.esults were achieved using two sleeves for the station-der. An outer sleeve made of copper gives an additional

freedom for the brass sleeve as well as absorbing heato avoid damaging the shoulder.l lifetime of the plastic stationary shoulders will be

using two sleeves around the pin.ting side of the weld suffers from lack of proper stirring

generation.o compensate the lack of heat by using stationary shoul-otational speed of the tool must be increased.

gments

rk was supported by the FCT project PTDC/EME-6/2010. Dr. Moreira acknowledges POPH QREN-.2 Promotion of scientic employment funded by theTES.

A, Azdast T, Doniavi A. An experimental study on mechanicals of friction stir welded ABS sheets. Mater Des 2013;43:4029,.doi.org/10.1016/j.matdes.2012.06.059.K. Materials ow behavior during friction stir welding of aluminum.99:229s37s.i K, Mazinani M. Forming nanocrystalline surface layers in cop-g friction stir processing. Mater Manuf Processes 2011;26:9225,.doi.org/10.1080/10426914.2011.564253.A, Raju LS. Inuence of tool pin proles on friction stir

of copper. Mater Manuf Processes 2012;27:14148,.doi.org/10.1080/10426914.2012.689455.A, Besharati Givi MK, Zareie Rajani HR. Experimental inves-of material ow and welding defects in friction stir weld-aluminum to brass. Mater Manuf Processes 2012;27:14028,.doi.org/10.1080/10426914.2012.663239.

S, et al. Analysis of effect of tool geometry on plastic ow dur-n stir spot welding using particle method. J Mater Process Technol:145563, http://dx.doi.org/10.1016/j.jmatprotec.2010.04.003.. Microscopic analysis of the morphology of seams in fric-

welded polypropylene. Express Polym Lett 2011;6:5462,.doi.org/10.3144/expresspolymlett.2012.6.Czigny T. Applicability of friction stir welding in polymeric. Period Polytech Mech Eng 2007;51:15, http://dx.doi.org/10.3311/07-1.02.n T. Fatigue of friction stir welds in aluminium alloys that con-

aws. Int J Fatigue 2003;25:1399409, http://dx.doi.org/10.1016/23(03)00060-4.

a AB, Fratini LG, Palmeri D. On microstructural phenomena occur-riction stir welding of aluminium alloys. J Mater Process Technol:3403, http://dx.doi.org/10.1016/j.jmatprotec.2006.03.192.N, et al. Effect of friction stir welding parameters on morphol-strength of acrylonitrile butadiene styrene plate welds. Mater Des45764, http://dx.doi.org/10.1016/j.matdes.2014.02.036.


[12] Rezgui MA, et al. Application of Taguchi approach to optimize frictionstir welding parameters of polyethylene. EPJ Web Conf 2010;6:07003,http://dx.doi.org/10.1051/epjconf/20100607003.

[13] Bilici MK, Ykler AI, Kurtulmus M. The optimization of welding parametersfor friction stir spot welding of high density polyethylene sheets. Mater Des2011;32:40749, http://dx.doi.org/10.1016/j.matdes.2011.03.014.

[14] Arici A, Sinmazc elyk T. Effects of double passes of the tool on frictionstir welding of polyethylene. J Mater Sci 2005;40:33136, http://dx.doi.org/10.1007/s10853-005-2709-x.

[15] Strand S. Joining plasticscan friction stir welding compete? Electrical insula-tion conference and electrical manufacturing & coil winding technology confer-ence, 2003. 2003. p. 3216, http://dx.doi.org/10.1109/EICEMC.2003.1247904.

[16] Strand SR. Effects of friction stir welding on polymer microstructure, MastersThesis, Brigham Young University, 2004.

[17] Scialpi A, et al. VibladeTM: friction stir welding for plastics. Weld Int2009;23:84655, http://dx.doi.org/10.1080/09507110902843271.

[18] Coeyrehourcq KL. Etude de mthodes danalyse rapides de la struc-tur molculaire du polythylne; 2003, http://dx.doi.org/10.1051/jphys:0198400450100160700.

[19] Ahmadi H, Mostafa Arab NB, Ashenai Ghasemi F. Optimization of processparameters for friction stir lap welding of carbon bre reinforced thermo-plastic composites by Taguchi method. J Mech Sci Technol 2014;28:27984,http://dx.doi.org/10.1007/s12206-013-0966-1.

[20] Yousefpour A, Hojjati M, Immarigeon J-P. Fusion bonding/welding ofthermoplastic composites. J Thermoplast Compos Mater 2004;17:30341,http://dx.doi.org/10.1177/0892705704045187.

[21] Rudolf R, et al. Welding of high-performance thermoplastic composites. PolymPolym Compos 1999;7:30915.

[22] Nelson T, Sorenson CD, Johns CJ. Friction stir welding of polymeric materials.US Patent 6,811,632, 2004.

[23] Mendes N, et al. Morphology and strength of acrylonitrile butadiene styrenewelds performed by robotic friction stir welding. Mater Des 2014;64:8190,http://dx.doi.org/10.1016/j.matdes.2012.06.059.

[24] Colegrove PA, Shercliff HR. 3-Dimensional CFD modelling of ow rounda threaded friction stir welding tool prole. J Mater Process Technol2005;169:3207, http://dx.doi.org/10.1016/j.jmatprotec.2010.04.003.

Shoulder design developments for FSW lap joints of dissimilar polymers1 Introduction2 Experimental procedure3 Results and discussion3.1 Conventional shoulder3.2 Stationary shoulder3.3 Probe3.4 Axial force3.5 Tensile strength

4 ConclusionAcknowledgmentsReferences

Documents

Shoulder desain .pdf