ASD TR-7-888 (I) ASD INTERIM REPORT 7-888 (I) September 1961

DEVELOPMENT OF ULTRASONIC WELDING EQUIPMENT FOR REFRACTORY METALS

J. Byron Jon" Nichola• Marapll

Comtlne F. De PriiCO John G. Thomcn

JOftet Devine

AEROPROJECTS INCORPORATED West Chester, Pennsylvania

Contract: AF33(600)-.43026 ASD ProJect No. 7-888

Interim Technical EnglnHring Report 1 June 1961 to 31 Augull 1961

Joining refractory and su.,:WOIIoy metal• in tl.kknesses up to 0.10 btch by ultrasonic weldt~~g Ia lecnf.b1e. A Ant. opptoxlmotlon of the acoustical energy requltM In~~ lhat tho requlslte equipment Ia also ftot1~41. Thl• report Ia a compilation and dfscunion of lnformOtlon pertinent to tflo •'Pfqpment of ultrasonic welding ~ for tolnlag AM-355 .e .. l, lnconol x,~• 41, tungstett, molybclonum­OS.'Jhl~, ~ftd coh•mblum alloy (duPont D-'3'fT-

FABRICATION BRANCH MANUFACTURING TECHNOLOGY LABORATORY

AFSC Aeronautical Systems Olvl•lon

United States Air Force

Wright-Patterton Air Force Base, Ohio

ASD TR ... 7-8S8(I) ASD INTERIM REPORT 7-888 {I) September 1961

DEVELOPMENT OF ULTRASONIC WEUlDIG EQUIPMENT FOR REFRACTORY METALS

J. Byron Jones Nicholas Maropis Carmine F. DePrisco John G. Thomas Janet Devine

44406

AEROPROJEC~S INCORPORATED West Chester, Pennsylvania

Contractt AF33{600)-43026 ASD Project No. 7-888

Inte~~u!:cm~a!o Enf":::!~gl;6iW A. R ~! ! ~·1 G The authc:nticit\' 0 .~o t'·,. ·. . h ·. · _ " ·" ',. r~·:'~ort as not been estadi~L ,:. l u h t list as source m~t--·tTJl 1 , . h

. . - • , Cd: I, r:·.'r r:,nroc~uce t lS T<·n -, .-t n•· t ,- • -, -" •n ,lout 'J ... ,.. . ... , h • • l: \...!.L,d.tV.:::.l,Jd or t e ISSumg agency.

~ining refractory and supe~alloy, metals in thicknesaYs\?~,~ ultrasonic welding is feasible. A first approximation of the acoustical energy- reqUired indicates that the. re<iuisite equipment is also feasible. This report is a compilation and ·discussion of intonnation pertinent to the development of ultrasonic-welding equipment for joining AK-355 steel, Inconel J:, Bene lalf tungsten, molybdenum-0.5% titanium, and columbium

alloy (dul'ont D-311.:)~ (9.l~'t

FABRICATION BRANCH MAHUFACTUR.ING TECHNCLOOY LABORATORY

AFSC Aeronautical Systems Division United States Air Force

iright-Pattersan Air Force Base, Ohio

A

A':~ ... ' I'HJ\ c·r -STl ·

ASD TR-7-nBS(I)

--··· Q__4U£

ASD Interim Report 7-888(I) September 1961

DEVELOPMENT OF ULTRASONIC WELDING EQUIPMENT FOR REFRACTORY MErAIS

J. Byron Jones et al

Aeroprojects Incorporated

lJoining refractory and superalloy metals in thicknesses up to 0.10 inch by ultrasonic welding is feasible. A first approximation of' the acoustical energy required indicates that the requisite welding equipment is also f'easibl~ This report is a compilation and discussion of' information pertinent to the development of' ultrasonic welding equip­ment for joining AM-3.5.5 steel, Inconel X, Rene !a., tungsten, molybdenum-0.5% titanium, and columbium alloy {duPont D-31).

(The feasibility of' welding the materials and gages of' interest is supported by data, appropriately referenced, from previous work with thinner material. Information on transducer, coupler, and tip materials is presented with information on evaluating efficiency and practicability. Design information on spot-type and roller-seam welding machine tips is presented. Data on various properties of' the we ldment materials are tabulated.:J . ·

~p6

ii

2

2

'

;a::muw:

FOR'I!WORD

This Interi:in Technical Progress Report coven the vork performed under Contract AF33{600-43026 frm 1 June 1961 to 31 August 1961. It is published for technical information only and does not necessarily represent the recommenda tiona, conclusions, or approval of the Air Force.

This contract with Aeroprojects Incorporated or West Chester, Pennsylvania, was initiated under ASD Manufacturing Technoloo Project 7-888, "Development of Ultrasonic Welding Equipment for Refractory Metals". It vas administered under the direction of Fred Miller {ASRCTF) of the Fabri­cation Branch, Manu.facturing Technology Laboratory, AFSC Aeronautical Systems Division, Wright-Patterson Air Force Base, Ohio.

The project is being coriducted under J. Byron Jones, Aeroprojects Director of Research; with Nicholas Maropis as the engineer in charge. Others tiho cooperated in the research and in the--preparation of this report are Carmine F. DePrisco, Chief Electronics Engineer; J. G. Thomas, Metal­lurgist; and Janet Denne, Physicist. This report has been given the Aeroprojects internal number of RR-61-75.

This is an interim report1 and the data reported herein are or a preliminary nature subject to analysis and modification as research progresses.

PUBLICATION REV:mw

Approved by:_.,)A't.J,~!12~~~W-J D

iii

4 a

,

I

II

Q.[ 2 , A 3

TAB!.E OF CONTENTS'

ABSTR.AC'.r • • • • • • • • • • • • • • • • • • • • • • • • • • • • 1i.

INTRODOOTION • • • • •••• • • • • • • • • • • • • • • • • • •

A. B. c.

• • • • • • • • • ••••• • • • •

Background • • • • • • • • • • • • • • • • • • • • • • • • Selection of Materials • • • • • • • • • • • • • • • • • • Properties and Other Pertinent Data of Materials •••••

WELDil«l ENERGY CONSIDERATIONS •••••••••••••••••

A. B. c. D. E. F.

Predicting W8ldability and Power Requirements • • • • • • • Clamping Force Determination • • • • • • • • • • • • • • • WSld-Zone Temperature Measurements • • • • • • • • • • • • Experimental Equipment • • • • • • • • • • • • • • • • • • Experimental Procedure • • • • • • • • • • • • • • • • • • Energy Requirements and Clamping Force • • • • • • • • • •

1

s s s

11

11 13 14 14 17 19

III ACOUSTICAL MATERIALS SURVEY • • • • • • • • • • • • • • • • • • 22

IV

v

A. B. c.

Transducer Materials • • • • • • • • • • • • • • • • • • • Coupler Materials • • • • • • • • • • • • • • • • • • • • • Tip Material • • • • • • • • • • • • • • • • • • • • • • •

ACOUSTICAL MATERIALS STUDY • • • • • • • • • • • • • • • • • • •

A. B. c.

Transducers • • • • • • • • • • Couplers • • • • • • • • • • • • Tips • • • • • • • • • • • • • •

• • • • • • • • • • • • • • • • • • • • • • • • • • • • • •

• •• • • • • • •

ENERGY DELIVERY METH(])S • • • • • • • • • • • • • • • • • • • •

A.. B.

Systems • • • • • • • • • • • • • • • • • • • • • • • ••• Components • • • • • • • • • • • • • • • • • • • • • • • •

'!'HI LIMITATION ON AMPLITUDE SET BY MAXIMUM S'l'RAIN ENERGY IN VIBRATING SYSTEMS • • ••••••••• • •

LIS'!' OF RBFERENCES • • • • • • • • • • • • • • • • • • • • • • •

iT

22 23 32

34

34 36 37

39

39 42

48

S4

4

..

.L _ ILL 1 H-%¥

LIST OF FIGURES

Page -1 Sketches of Typical mtrasonic Welding Systems ••• • • • • • 3

2 Variation of Weld-Zone Temperature at Different Clamping Forces and With Elapsed 'lime During Welding • • • • • • • • • 21

3 Calorimetric 'lest Scheme for Transducer and Coupler Evaluation 35

4 Transducer Designs Incorporating Ceramics • o • • • • • • • • 43

LIST OF TABLES

Table

1 Selected Physical Properties of W6ldment Xaterials • • • • • • 6

2 Selected :Mechanical Properties of Weldment Xaterials • • • • • 7

3 Xetallurgical Properties and Anticipated Weld-Zone Temperatures of Various Weldment Materials • • • • • • • • • • • • o • • • 8

4 Experimental Welding Data and Predicted Energy Requirements for Selected Weldment Xaterials • • • • • • • • • • • • • o • 10

Estimated .Acoustical Energy and Power Required to Weld Xa-terials 0.1-Inch Thick in Annealed or Stress-Relieved Condition 12

6 Comparison of Melting Points and Recrystallization Temperatures

7

8

9

10

11

12

or Four Refractory Metals • • • • • • • • • 0 • • • • • • • lS

:Material Ordered For Phase I ••• o ••• o • • ••••••

Estimated Welding Energy For Available Gages of Materials of Phase I Investigation • • • o • o • • • • • • • o o • • • •

Data on Surface Inspection of Specimens of Four Katerials •

Properties or Transducer :Materials • e • • e • e 0 • • 0 • e

16

18

20

Physical and Mechanical Properties {at Room Temperature) of Candidate Coupler Materials • • • • • • • • • • • • • • • • • 26

Acoustic Properties or Candidate Coupler Materials • • • • • 27

v

UIS%1

y

M .J, '):-,

LIST OF TABlES (Concluded)

fable Page

13 Machining and Joining Characteristics of Candidate Coupler Materials • • • .. • • • • • • • .. • • • • • • • • • • • • • • 28

14

15

16

Impedance Matching Between Candidate Coupler and Transducer Materials • • • • • • • " • • • • • • • • • • • • • "• • • • •

Typical Physical and Mechanical Properties of Candidate Tip Materials • • • • • ., • • • • • • • • • • • • • • • • • • • •

Relationship Between Relative Strain Energy Levels For Con­stant .Amplitude and Relative Amplitude For Constant Strain Energy- Levels • • • • • • • • • • • • o • • • • • • • • • • •

17 Characteristics of Terminal Tip-Reed Arrays For Single-Spot Welders •••• • • • • • • • • • • • • • • • • •••••••

29

31

40

18 Characteristics of Terminal Tip-Coupler Arrays for Continuous-Seam Roller Welders • • • • • .. • • • • • • • • • • • • • • • 47

!he increasing use or the newer, high-t•perature, corresion­resistant aetal!a and allo18 in Dlissile, &pace Tehiole, and_ atOJdo applica­tions has intraduoed. new J18tal-jo:!n11\f: probleas tliat can not be readil;r SOlTed Dy C0nv8Dtioul teClmiquas •. hoduc!q satiaractor;r 'bcm.da iJl Suoh materials as ao1Jbd,el!l\1D1 :R,ene 41, .tnconel I, and AH-355, in both aiaUar and' disa:lmUa:r caabinationa ot medium. and heaV7 gages, present certain dittieul ties.

Since the ruSt wchnical paper on ultrasonic veldiD.g (l)tt vas publiShed, this aubje~ has nJC81Ted fncreasing ~tteution at TarioUa •tal­lurgioal conterences (2-6), as well as from .&;merioan industr;r (7-1$), espeeiall;r the Jll8tal fabrication industr;r (16-23), and rroa foreign inves­tigators (2h-29,).

Ultrasonic 118lding equip!lell\. alread:r deTe1eped aac1 im use has d8monstrate4 i \a effectiveness in joi:lliRg Tarious aateriala or interest t.o the aerospaoe induatries. OO;r in some or the al'UIIbla. allo111, however i

• has 118ld1Dg 'been possibie ill the heavier sheet gages (up te about 0.090 iJl~). Wi~h ~- ~itiDg eqdpnt, the ga1e r.,_~ :aost other uterials is IiJd\ed \o about o.()ho tach. .Ert;ension ot the utUit7 or the process to heavier aad harder materlals requit-es substantial increases in t.he •t rl­'b,rator:r powar deliTered. to the wid lon&o SUch i.Doreasea can COJI8 Tia 0Dl.7 t1f'o &T8BU8S I

,

~:r- .. .!~::r::;:::l!:f;:st:~!.~r:~;~i:no:t~~~ane:=:!!: cou.pliilg syatema

2. iaoreasecl poW'er to ·the tranad.ucer-couplirlg &78ta.

!lie major objectd.Te o? Phase I or this program is to develop ultr·­sonie vetd.i:q eQu.ipment adeqaate ror .3oining_1;ha harder~. higher strenath.•tala IUld .allOTs h thicknesses up to about 0.10 b.ch.. ·'l'o achieve this o'bjecti"Te it wil~ \)e _nec.•••&rJ to. establi•h the. :te~sibllit,. . ot _jo.ining metallic materials, as ex~litied. b7 coluabima allOT,· aolJbdantml (Ko-o.S' Ti) allo7, tunga\en1 Renl U, AK-3$'), and Inconel I, b. aonometallic and dissimilar uterial coa­biriatf.ou 8nd t.o .Utline a a;yat9Jilat1.o ap:PJ:-oacb to the dsvel.,..nt of teCh.;. niqlleS a:nd equipll8Ut D8CeB8&1"7 tO lll8.ke :reliable 1 reproducible SeU. and &pOt­type ultrasonic welda.

"luahera in parentheses reter to :r.terences lis\ed at end or reporl.

1

a

•

XC .& .L,L&

htera!Dation ot equipment requir•ents for uliorascaicall7 weidiD& metall1c m.teriais iB a iipec1rio thickness range oat begin with. a stuq of · the &DeJV' requ:ireaents for uld.ng the wlu. This is .not a utter .of •rely detilrl.Dg the }iDe powr r•utaired to operate welding eqll!.pment, nor does it deal solely With the more e..Un: probl811 Of the acoustical energ' d.eliTered into the wld. zone. Actuall7.,. the :tl.o1r of energy through the entire electro­acoustical s.ystem must be considered.

·· ltiectrical power from a standard power line ( 60 cycles) is d,eliTered into th.e "Ultrasonic generator• or power source 1 where it is co:DV'erted by aeans of au1l.iary electrical eqaipaeat 1 such as electronic oscillators and. power ampiitiers, into eieetriCal power at the eperating trequency or the wldiac JU.Ohi.Da. '!'his high-trequenc;r electrical power .. is delivered. to ihe transducer, .which con.Verts it ln:tio vibrato!'T power of the same frequency. ·The J)Olf8r then passes through the coupling s;rstem, which u;r consist of one or more lllelllbers, into the weld.ing tip and the Etal being joined.

a'ertain elements are common to transducer-coupling s;rsteas tor welding, and these require development for effective use in. higher power u1 irasonic welding machines.. Transducer material ma;r be .. selected from a variety or candidates, and transducer designs depend in large measure on the selected transducer material. Coupling material 1mst be selected with con­sideration o~ certain material.preperties, some or· whiCh ma;r not haTe been quantitatiTel;r established. 'Welding machine tips involve especiall;r diffi­cult requ.i.relitents.

The basic concepts ot these s;ystems consist of a transducer, a. coupling STStea, a weld:illg tip, and an anvil or npport tor the workpiece. Alter. the most promising s;rstea elaeD.ts are de.terained, the best potential coupling system nat be selected. from two general classes and a TarietJ' of types.

·:rn tb8 wdge-fted s;rsta, used in hiiher power apo.,.type WEU.d&ra, acoustical ensrgy is deliTered to a wedge-shaped lll81Der (a mechanical tr&D8-toraer) which e:x:ecu:tes longitud.i:nai vibration and excites the reed member in flezural vibration at a SGI18what greater amplitude than 1& produced by the tra.DIId.ucer, ea:a.aing the weld.i.Dg tip to vibrate esseatial17 parallel to the weld interlace.

·· sUn•r welders ~d. . portable-t;ype welders . eonve.nientiT incorporate the lateral-drive s111te of FiS• 1. In this case, the tip is attached to a coupler which vibrates longitudina.U7. to produce tip exeursion parallel. to the weld ::Lntertaoe. Cl&llping force is applied through benclf&tg of the coupler •

. A ring-weldiag.ll&ehine ia essentiall7 a special Jdnd of spot-type welder that produces an uninterrupted annular weld With a single 1 short power interYal• SUch a welder utilizes a torsionally d.riTen coupler S)'"Btem. I:n.

2

•

I

~ .. A~~-J F'ORCE

A. WEDGE~ED SYSTEM

MASS

TRANSDUCER

WELDMENT

C. CO:m'INUOUS-SEAM SYSTEM

L

B. LATERAL-DRIVE SYSTEM

1 FORCE

DC POLARIZATION

COUPLING /SYSTEM

ANVIL

AC POWER

D. RING WELDER

Fig. 1: sm'CHES OF TYPICAL ULTRASONIC WELDING SYSTEMS

3

•

one tTPB o:t ring welder arrangeaent, illustrated in rig. 1, the longitudiD.al17 vibratiag "horns• (•chanical \ranatormers) are attached. approx:lmatel7 tange11t to the torsloJIIll. reed .. ber, produoiDg .. torsional dia.pl~oements or the welding tip. Other arrangements tor producing this torsional Tibration have also been developed •

.l contiimOUS-Ie8Jil welder incorporates a lateral-driTe transducer­couplinl &78t8JI rotating on antitrietion beariags with power iatroduce4 th;-ough slip rings, usuall7. with Jt!'QTiaion tor rotatiol'l or the eat~ ~rana­duc8r-couplillc df.U:.;.tip B:flltD b7 a motor drive. ,J. disk or ri:Jis·liD tip operates b. IQ!iehrolloa.a rolliq 9ont~oc;t with the work so that there is essen­t.iaii.T ao aiippage beilwen the tip andl the work •

a 434~. ;__,

I" Ml'J.'IB.fi.L lliiDIIG .J'U.SIBILift

Es\4l'ilish tbe teasi.bUit7 ot joining retractoey metals and ot joini.Dg a retracto!7 •tal· to a dissblilar design material b7 ul truonic techniques.

A. Ba!kgroUDd.

~"'"!his work is aoncemecl with shewing the teadld.lit7 ot 11ltrasGnic~l7 welding Sll.ch retractor,- .•tala as tungsteD, ao1Jbdell•1 tantal:aa, _aDd eoluhia a:nd. ncb other desiga. aater1aJ.s as the superallO'J'S tJPitied, bJ' AJI-35) steel 8nd tJdiaet 70o.J bBe aaterials are reiatiTel7 ..... , aDd their properties are not as wil ci.nb.ed. as those of sucb aore coa.on 118tals and allo;ys_ as al..mum., copper, and nicbl •.. considaratioa wil1, theref'ore, be given to a liaited mm­ber of materials tar COIIlprehensiTe stud.,- in the course or this program.

B.. fSe1s ction of Material~

. · Mmutacturing Technology personnel ot . t.S. Ae.r~u~ieal ~teas Db·ision of the Air rorca Sys.t•s C--.nd have reaom.endea the tollO'Iri:ag sb: materials. tor ·the focus of et.lorts during this progrmt

. .rs-. -!::3SS steel

cone! X l!),'e sn; Iii /1, ·~

c. Properties· and other PertiDeat. Data o~ Xaterials (30-lt.l)

1f1. th a Tift to. titti.J1g these speeitic materials iDto existiBa theory relardiag vltr••or4ct welding, as ,..11 as to 1is1isti.Jig in retiDial neh tlm0171 datlt, on 'ihe 'ph711ical properiies of· illese materials haft beea . .

.. ~f, ... b~ecl ip T~ble 1., uta oii the aeCJwdeal. prOperties are given ill !altle 1, . ed oe:rtaili lletallvci.Oal. uta are repor\84 ill fable ·3. Actditioul data will be incorporated iato the tables as the prop'&JI preceed.s.

G\

'table 1

Hu1t1- Temper-plie~- at~re· Rene ·u

Densit7~ ·lb/in~3- RoCR. 0.296

Linear Coefficient ot Ther- 10-6 Jtooa 6.5 Jlal Expansion, in./in.-•r J.ooo•:r 1.S

Thermal coad.uotlvi\71 RoCII 63 Btu-iD./tt2-~•:r iooo•:r lOS

-.

bnal. Diftuint.7, t\2/br ltoca 0.()95 1eoo•:r o.1S8

Speciiic Rea-\, Btu/lb-•:r Roca 0~108 lOOQ•F

. WeldEn\ Raterial and ConditiODit

.Ho-o~S'ri AM-3SS Inccmel I VJ.C-Sit Tungsten SCT -m 0.368 0.697 0.282' 0.298

3.1 2.6 6.4 . 7.6 3.2 2.7 7.2 1·1

936 liS 1~ 85 8kO 90 lh4 lo6

2.01 0.249 o.lk8 0.131 l..TS

o.o6i 0.032 0.120 0.10$ o.o63

D-31

0.292

h..l

• SO'fi subzero-coole4 and temj)eredJ SH'l't solution heat-treated; and VACh vacuUll are-east. Allmawrial procured in the armealed or stress-~lie.t-annealed condition.

itt!Coapute each ita or data with the aul.tiplier illdicated tor the propert7.

I

Table 2

-~~-~~JllO.~T~ !'Jf ~~- ~TDIALS

ui:t:hi.ate ·!ensile Strengtll, psi

Yie1dStreDgth {6.2% offset}, psi

JD.ongation, %

Poissons Ratio

Modulus· or· Elas­ticity, pa~

Shear Modulus,. psi

Hui ti- Temper-plierH •ture · len1f U

Rooa 16S,ooo 1000•F 1781000

Ro• 1000.,

140,000 134,000

20 13

6.310 o.32S

31.6 27.3

12.1 10.2

Weldaent lfaterial and Conditi~ Mo-O.~i jK-3SS IncQnel I . VM-SR Tungsten SOT SHT

lliS,ooo 12o;ooo 223_.~ 16o,ooo 110,000 7S,ooo 197,000 14o,ooo ·11s,ooo 100,000

4S.S

18,ooO

0

S9.o ss.o 21.8

19S,ooo 140,000

10 1

0.276

28.7 24.0

11 •• 9.4

120,000 83,000

25 10

0.290

31.0 25.0

12.0

D-31 100,000 '68,000

9.8,000. 68,000

1) s

0.380

16.S 12.8 6.0

• ·sB'h solution heat-treated; SC'fr . subzero-cooled and tempered; and VAC 1 vacuum arc-cast. All ~at~ri&l was procured in the ~aled or stress-relief-anne&led condition.

-Hcompute eaeh 1 tea of data vi th the mul tipller indicated for the property.

a ".&&JL~

Table 3

METJil.tRoiCAL JiROPERT:iEs AID AITiatPJ.TED WELD-zm ~'!'tiRES

Material Conditiort*

Crystal Structure

Reeryst8lli$ation Temperature, ep

Melting POint, •r

Jntieipate.d T.-pera~ ture fn W&ld Zone, •r

M'axi­Min:IIIUII.

~ VJltiOUS m,JJfUit MA.'.l'l!RIAtS

Mo-o.Sfi vm··

bee

2100

4370

Tungsten

bee

23)~2750

6170

AH-3>S SC!

**

25'00

1Q20 sao

Itic'onei ··x Slfr

**

2540

lOJ.Q s~o

•31

bee

1800-2100

4100

1820 ii35

it SO! a subze~c~oled .. ·abd. 'tf.predJ . Slftl ~6lution heat-treate4J and TACt vaouua az.oc--:~~t.· .lll material was procured in the a:imealed ar stress-reliet-annealecl condition.

HMu1 tiphaae structure J l!ltl"UctUr. of the ma tri% in the annealed coDdi tion · is f'cc.

8

,» a:

.ls will be shown later in this report, current theories of ultra­sonic welding relate the enera requir•ent associated. with producing spot­type welds to . the hardness and thickness. of the welct.nt. ma~r.~· Perti­nent data on the weldin&-eneru requ.iraa~nts ~or the six wldlleDt ll&terials of interest 11'1 vmous thin gages!'""" assembled trom the data of previous experlmental. work and are reported in !able b..

ATauable ultrasonic weldi.Dg d&ta indicate that although meticu­lous attention ·.to surface preparation is not necess&r7, oxide~ tree. and. dB­greased. surfaces respond. more read117 to welding. J.ccordingl71 letters requesting Wol"ll&tion on surface tUms and. the,ir propel"ties, as well as cleaning and. surface preparation procedUres, haTe been sent to the research departDaent of . each manutactlU'tJr from·llhca metals tor this prosram were purchased. To date, .five replies have been rec.eived. Additional requests are being tranBilitted to·· other material :l.ntormation sources.

inasmuch as .re~rystallization ot refractor;r.JD.etals and super­allo;ys. results in strength 'd.4t8radation o:r· such materials, recrystallization should be avo.ided. . Qle notmrth7 •ciVantage of ultrasonic welding is. the absence of a cast structure and, except iri unusual cases' of recr;ystalli­zation.

'R8cent ·research shows that the temperature rise coDIIlOnl7 observed in ultrasonic welds is in the range of 3S~-SO~ ot the homologous melting temperature. In .m~st cases, ~his· is below the temperature at which re­crystallization ti.~s place •... other rese!f.l"Ch (42) .. ahovs that temperatures during veldirig can be controlled within limits that are probabl7 adequate to preclude recrystallization.

!he reeeyBtailisation tamperatut'U., d.Ten in Table ;, represent a ftmlaa.ry ot publ11he4 data and · ot ~.otri. •1cl-aone t.tap8ratUr.a. !b.us, ~th delineation of suitab1e .welding maChine settings, the avoidance of recrystallization appears to be practical.

9

a

Table 4 ~At WELDING DATA AtnfPREb:CCTED···.-y RBQtJIRPDITS

Fcft S!:r.Jmm, ~ M.JmtJ!.s ·

,_

Prniaa.a Experience "

PrediCted. Material '!'rPical Weld. Energy

and. Energ7, Ciapiq Tensile Required, Hardiiese. ' Gagel 1Mb vatt ... aec roree, lb Strerigth, lb vatt-aec

Rene hl o.m.o 1000 800 3.50-SOO .oo6:-.o2o 26o

Mo-o.s 11 .oos· 720 .300 60 600-900 VMR ~ 26.5 .010 1200 .400 1200-16,50 to300

.ois 2000 400 220 2000-2800 600 100

2.500 6oo lSO .017 3000 600 2.50 2000..2800

~·tev.. .oos ~ lSO i6 900 VIII • .300 800 90

900 100 .010 2600 900 .1$ i6.30

1920 5oo S7

.:...j5~ .ooa l6Q .3SO 380 202: .3SO ~6() .3.50

rnconel·:t .012 soo-1000 100 207 610 vMH • 1.3S 45~1000 lSO 267

.020 lSOO tSo 290 1260

.0.32 1520

D-31 .. .C>o6 1200 3SO .38 700 VMH ,a 238 o.oos jooo 700 900

. *Vicbra micro indentation hardnesa n111lber.

10

a ~ 2 - J

n. WELDIIG BIERGY CONsl:DzRA:fiONS ~

Study the energ requirements for welding colUlllbium (D-31) alloy, tungsten,, mol)'bdenum-o.s• titanium Blloy, Remf bl, AM-3SS, and Inconel x.

The~ eneru requirements tor vibratory welding a variety of •­teriala, including sc:ae ~of the refractory metals, haTe been studied exten­sivel:r dllring the past sewral years. Appropriate equiplll8nt, te~iques, and instrllll811tation were deTeloped tor~ identit,.I.ng the nriou critical £actors associated .with 11ltrasonic _weldills eurgy and tor meaBUring such factors, including temperatttres in the weld~~ zone (42-44} • Experience and intorm.ation were acc1Dl\llated tor welding thin ~gages of retractorr metals (such as columbia, molybd.enu, tantal\Uil1 .and t'Uilgstf!n) and ~ superallo;ys (such as 17-7 m, .AK-3SS, J-lSOO, Inconel x, and Rene 41}.

~ . Since many ot the probl8118 encountered in welding sueh uterials haTe been recognized and Tariously, though not coapletel:r, solTed,, refine­ment and extension of this earlier work to the heavier gages ot the candi­date materials are needed.

A· Predicting Weldabilitz and POtier Requirements

an ~the basis of earlier fundamental ul trason1c-we]41ng research, a :tirst-appro:d.mation criterion tor determining the weldability of a given material in terms of thiclm:ess and acoustical eneru was postulated and defined b,- the equation {4S) 1 ~

I .-It ·a3/2 t3/2

where E is the acoustical enera in joules~~ ( -.-t;t~~econds), H is the V~ickers llicroindentation hardness number of the •terial, , t ~ is the thickness of the sheet in inChes,~ ~ It is a linear~ constant which in­corporates other contribut:lng Tariables.

since this enera equation was init.iall,- derived trca experiaen­tal data ~ obtailied over a period of time tor both common and exotic materials, predicting the wldability of a new Jll&terial by this means has proved. to be reasODabl,- accurate within the thiclmes• range so far studied. J'or thicker aterials, howTer, moditication of the equation may be necessary •. EDergr Nquirements tor welding up to 0.1-inch-thick refractory material were calcu­lated and are given hi,Colmm 3 ot Table ~. !he relevant acoustical power at T&rious welding intervals is giTen 1n Col'Dils 4 through 6 of the ta'ble.

n

Table 5

ISTIMlTD .lCOUSTICI.L EDIGY AID POWBR RIQ'DIRJ'J)It

TO BID lUTIR.IALS 0.1-IEB THICK II Am.AL'!D al

STRISs-RELIEVED CONDI'fiCJi

Power Required, tv

Material. R- 41 Mo-0.5 'ri

Tungsten

VJIHH-

300

265

;oo

lnerg Required, tv-sec

10.3

9.2

~0.3

At At At 0.1 0.5 1.0 sec sec sec -103 21 10

92 18 9.2

103 2l 10

AM-355 - - - - - -Not yet available- - - - -

Inconel X 29 5.8

D-31 236 1.5 15 15 7.5

* Based on equationr B • K a3/t t37'l. **Vickers microindentation hardness number.

12

a

Under certain circumstances, very short welding intervals may be mandator.y. iben a material has a ductile range at a somewhat elevated tem­perature, however, the desired result can sometimes be produced with a

. fair1'1long interval (such as 1 second) at low power preceding a second short interval at high power.

Considerable data, as well as experience, are required before the power requirements for the materials of interest in this program can be firmly established. As is evident, the power required of the welding equip­ment will be high when the requisite welding interval is short.

B. Clamping Force Determination

Experience has shown that the minimum power required to produce a good weld is associated with a clamping force which, within certain limits, permits the best iapedan~e match with the weldment. techniques that haw been developed for establishing the best clamping force are described in the following subsections.

1. Threshold-curve or lu.gget•Pullout Method

Clamping force req.uirement s for reasonably :aalleable materials are established by a method based on weld evaluation by a peel test {42, 46}. Welds are made at one clamping force and one welding interval but at de­creasing power as loag ~s the welds fail the pee:t test by nugget pullout. This procedure is repeated at various clamping forces to obtain data fer establishing a power1clamping force curve. '!'he min:l:aum of the curve corre­sponds to the optimum clamping force. '!'his curve may also be used as a threshold curve for welding as it indicates the threshold power, or mini-mum energy conditions OMEc), far welding.

2. 'thermal-Response Method.

For brittle materials, the nugget-pullout test is n()t feasible, so the thermal response (temperature in the ~ld zone, itself} is used to establish clamping force requirements. For a fixe~ p~r _setting the temperature in the weld zone, irrespective of weld quality, ~ approxi­mately maximum at the clamping farce associate~ with the mini.Jama of the pawertclamping force_ curve. T!lus, for brittle materials, a convex upward curve of tanperaturetclaping force can be obtained. With this thermal­response method, it is necessary to ascertain the powar setting required to produce a weld at each clamping force 1 whereas with the nugget-plil.lout method., the power value is obtained at the s8lll8 t:tme as is the clamping force level.

a t _.J!liSJBL *

3. Standing-Wave-Ratio Method

The power delivered by any u1 trasonic transducer-coupling system. can be monitored by observing the stan.diDg elastic wave ratio existent on the coupler; a standing-wave-ratio method is used to establish clamping-force values at a fixed power setting (42). Microphone-type elements are used to detect the standing-wave pattern along the tran811litting system and to measure the ratio o! ma:rlmumtminim'Uil particle displacement along the acous­tic coupler (the associated standing-wave ratio). This is accomplished by applying the electrical signals derived !rem the microphone elements to the vertical and horizontal defiection plates o! an oscilloscope; the result is a varying elliptical pattern, the area o! which is proportional to the mechanical power passing through the instrumented portion o! the coupler at any instant.

There is also a direct relationship between the thickness o! a material and the clamping force necessary to produce an ultrasonic weld at a minimum power level (42). With the clamping force .established !or a specific thickness o! a given material, the best impedance match between the sonotrode tip and the weldment can be obtained so that delivery o! energy into the weld area is maximized.

c. Weld-Zone Temperature Measurements

Elevated temperatures in the weld zone may cause recrystallization with consequent degradation o! the weldment material. Reliable methods have been developed !or measuring weld-zone temperature rise . during vibra­tory welding by meltable-insert and precision single-wire the.rmocouple insert techniques. Examination o! the earlier eXJ)erimental. data thus ob­tained (46) shows that the temperatures obtained at the weld interface are ordinarily only 3'-50% of the absolute melting point ot the weldment ma­terial. rn general, the recrystallization o! a metal depends upon its prior conditioning, so the recrystallization temperatures given in Table 6 are intended as representative approximations expressed as percentages o! absolute melting points. Although the temperatures shown in the table are at the high end of the 35-50% homologous temperature range, associated with u1 tr~sonic welding, recrystallization or the weldJnent usually can be avoided by welding at the minim.um energy condition.

D. J!Perimental Equipment

The welding energy requirements !or each gage or each material listed in Table 7 will be calculated !ram the previous1y discussed easrgy equation when the Vickers microind.entation hardness nUilbers have been established !or all o! the materials.

'!'able 6

Ca!PARISOI or M!LmG POINTS AID RICRIS'l'.ALtiD:TIOI

TEMPJ!BA'l''llii!S OF FOUR RlmllC'l'OlY MI'1'AL8

Malting Point Recr.ystallisation TaRperature % ot Absolute

Jlaterial. •o ., •c •r Mel tina Point - -Mo-O.S fl. 262S 2898 1100 1373 47

'l'ungstea 34m 3683 1400 1673 47

D-31 A1.loy 24lS 2688 1000 1273 47

'l'antallDI 2996 3269 1300 1573 48

'!'able 7

MA'l'ERIAL* ORDERED FOR PHASE I

'!'hick- Quan-ness, t;:~·

Material Source iilcll

Rene hl Hamilton Watch Co. 0.008 2 · .• 02oH .0301Hf .o40H

Mo-0.5 Ti UbiTeraal Cyclops .010 3 ~ .015 3

.020 1

.030 1

'l'lmgsten J,i'anateel Metal- .010 2 lurgical Corp. .015 1

.020 1

.030 1

AM-35SH Source tor thiclmesa range ot interest and 8JII&ll quantity required not ,.t located.

rnconel X Whitehead Metals .010 1 .020 2 .031 1 .043 1

D-31 Alloy E. I:. dUPont .010 3 delfem.ours &. Co. .015 .1-1/2

0.025 1

* m material . procured .in the annealed and/ or streaa-reliet-annealed condition.

HJlot on hand aa o! 15 September 1961.

16

Q2 £U "2 AX

For each material thickness and cambinations thereof, a best im­pedance match into the weldment w111 be established by the minimum energy condition approach. Iri order to extend the weldable thickness for each weldment material to the lillit of the presently available equipment or of the proposed ju.ry-rigged equipnent, a thickness range for each material was estimated on the basis of previous work and on values computed from the energy equation (!able 8).

E. !xJ>!rilD.ental. Procedure

The experilllental. work follows a well-established pattern beginning with the preparation of materials and proceeciing through the determination

· ot welding machine settings, generation ot welda, and evaluation of results.

1. Material Inspection and Preparation

Inspection of materials for ·cracks and other imperfections• which are likely t.o interfere with the welding process, i1 particularly iaportant when materil.ls are brl ttle and, therefore, more SlUICeptible to cracldng. Inspection tor surface oil, I!Cale, or oxide, which must be removed 'by suita­ble chemical or mechanical means prior to weld.:il1g, is also important.

2. Welding Machine Settings·

The requisite welding energy, e"'tiJuted by means of the energy equation,. is used to adjust the power level for welding at weld intervals ranging from 1/2 to 1 second •

... Depending on the properties ot the welclaent mterial, clamping force at Jdn:iJilUil energy condition is ascertained by one of tbree methods: threshold. cunes, thermal response (EMF),· or standing nve ratio (SWR). '!'be first 11ethG4 is satiatactor:r for thin,. ductile materials but · MDnet be used with brittle or heavier gage materials. In the present work, with the refractory materials, both the other two methods are satistactory tor estab­lishing clamping-force levels.

3. ~ld !Valuations

~e weld characteristics are. evaluated on tbe basis ot one or more ot the following methodst tenaUe-shear. stre:11gth deter.ld.nation, cross­tension at~Dgth tests .. , microac;opic surface in.P.ction, x-ray inspection ot the subsurface, and metallographic study of the weld section and/or pro­gressive planar section~

~·

•

Table 8

ESTDIA.'l'ED WELDING, ENERGY* FCft AVAit..A.Bi:E GA:GES ~ MATER.m Fat. PHASE I IHVESTIGAfiOR

Average Jfelding· VMII*"* Gage, In;:,...,

KaterialH Value inch kw sec

Rene 41 300 0.008 0.3 .020 1.0 .040 2.0

iro-o.s Ti 26S .010 .3 .ms .6 .020 1.0 .030 2.0

fungaten 300 .010 .3 ~015 .6 .020 i.o .030 2.0

IX.;3.).) Information not avaUable at this t:t.

Inconel I 13S .020 .3 .031 .6 .043 1.0

D-31 Allq 240 .010 .,; '

.• OlS 6' ' .

0.02, 1.0

* Calculated trOll energy equation. HAll materials procured in annealed

and/or atreaa-reliet-annealed condition. *"*Vicker 1 s micro indentation hardness

number.

18

F. Energy Requirements and Clamping Force

The experimental work tor the present program was iaitt.ted with O.Olo-inm tungsten. 'l'he surface inspection t~ this material taUed to reYeal any serious i.mPertections (Table 9), and the test .spec:lmens were deg"ased llith ;Pennsalt A-27 cleaner ~ior to welding. J. laboratory-type instraented welder, acc011111lodatin& a. standard. 2-kw wedge-reed transducer­coupling system and a standard reaction anvil, was used in these first tests.

. lrom the energy equation, about )00 watt-seconds was estimated as the enera required to weld the o.olo-in~ material. Accordingly, in scouting tests to establish the. proper clamping-force level, the input power was arbitrarily adjusted to somewhat o"f'er 300 watts and the weld interval was set at 1 second.

Because ot the brittle nature ot tungsten, the clamping force tor this material was determined by both the weld-zone temperature (by single tine wire, 3-mil. Constantan theraocoa.ple} and the standing .... ve­ratio techniques J a t;ypical wld-intertace-temperat~·protUe as shown in Fig. 2A, was obtained with ~ sillgle-wire theraoooaple located approxi­mately in the center ot the wld. Th.ermal values ~ plotted in Fig. 2B, tor clamping forces 1n the range of 100 to 750 pol.Ulde-the maxiaul tempera­ture corresponds to a clamping force ot 250 polUld8. With the standing­wave-ratio technique, a cl.a.mping force ot about 300 pounds was indicated by the ellipse area depicted on the oscilloscope. Qn the basis ot these two sets ot measurements, the best clamping force tor o.olo-inCh tungsten appears to be 250 to 300 pounds.

19

!able 9

DA!J OB StJIUP!CB II'Sl'm!IOB OF Sl"!!'niMEIS OF FOUR XAft:Rnts

General surface over all test speci­mens was tlat, except tor D-31 speci­mens which had wavy surface.

Surface Gage, Roughness,

Material 1DMl. microinchee

Mo-o • .; !1 0.010 15! 2

Tungsten .011 30 = 3 .015 15 ~ 1 .020 20 = 3 .032 38 = 3

Inconel I .020 5~1 .03) 20. 2 -.040 25! 3

D-.31 .005 12! 3 .010 6 •1 .015 li .. i -

0.020 4 + 1 -

20

1000

800

p ~

j e600

!

1000

800

I I I

I I

I I

l .Acoustic Power t 300 watts Weld:l:nc InterVal. r 1.0 sec

200 00 600 800 Clam.piDg Jorce, lb

At VARIATioNS OF 'tfELD•ZONE TEMPIRA'rtJRE !'l DI:mCRDT CLAMPDG Ft'ECES'

Acoustic Power1 300 watts Welding :rnterYall 1.0 sec Clamping Forcet 200 lb

o.Js. o.~ o.6 0.7 o.8 0.9 1.0 Elapsed Time from Start of Welding, sec

B 1 VARIATIOI OF WJZD-ZONE TEMPERATURE wim IL.A.PSED TIME DURING WELDIBG

:ric• 21 V-ARIATION O.F WEID•ZONE 'l'EMPERA'rtJRE AT DIFFERENT CLAMPING Fm.cES JJID WITH ELAPSED· 'liME DuRING WELDING

21

LL£2

m. ACOUSTICAL HA'l'ERL\LS SURVEY

•survey o£ CUX1"ent and projected state-of-art materials £or their application as transducers and associated equipment with the ob­jective of delivering sufficient power to join the selected ma­terials in thicknesses up to 0.100 inch.•

.A. Transducer Materials (30, 47-58)

· "Transducers and associated equipment • embrace the entire electro­acoustical system £rom the connections tor electrical eneru input to the point ot vibratort energy output, the locale where the transducer-coupling system contacts the area of weld generation. ·

Durin& recent ,ears, a wide variety of magnetostrictive materials have been eTal.ua:te~ and used in e~rimental and produetion-tJP& ultrasonic welding arra)"'J these include 2'1 Permendur, Alfenol, •A• nickel, and nickel ... cobalt {2<il.) alloy. Such materials have a lower ettic~ency than some. ot · the electrostrictive ceramics; however, with metallurgical methods such as brazing, rugged and durable systems that are relatively insensitive to overloading can be bull t.

Furthermore, such systems can be operated w1 thout permanent damage at temperatures much higher than could be tolerated by any ceramic available until recently. Nickel has been the most effective and widely used of the magnetostrictive materials. Most ultrasonic Welding equipment incorporates laminated stacks of thin, annealed "A" nickel sheets which are satistactor.y tor heavy-duty, continuous operation.

The electrostrictive barium titanate ceramic, currently used in certain types of ultrasonic eqa.ipment 1 dates back to about 19$0 when the material was extensively investigated and used tn ultrasonic arra78 for solid-state metal treatment. Since that time 1 barium titanate has been used 1ri ultrasonic arrays for various purposes. Whlle its electrcmechani­cal conversion efficiency is higher than th&t of magnetostrictive materials, it has not been used extensively in production-type ultrasonic welding equiJa&Dt because ceramic transducers of this type are fragUe and some­what difficult to install in coupling systems on a practical basis. Furthermore, its low Curie point (approximately ns•c) introduces a major cooling problem ~ overheating must be prevented to avoid depolarization.

22

Recently, effort has been directed toward the development o.f new ceramic materials which will withstand high temperatures. These newer ceramics include such .family groups as titanates, niobat~s, tantalates, and zirconates. One o.f the most promising o.f the new materials is lead zirconate titanate, which has a reported Curie tempe:rature or about 34o•c and a high electromechanical coupling coetficient. Large-size transducers have been .fabricated .from this material (designated Brush Type PZT-4 and PZT-5) and evaluated.

In order to bring the transducer problems into sharp .focus, avail­able data on magnetostrictive and electrostrictive types have been compiled or calculated and are summarized in Table 10. Of' illlmediate interest, is the electromechanical coupling coefficient, the reported power-handling capacity, Curie temperature, thermal conductivity, and di.f.fusivity data.

The electromechanical coupling coefficient serves as an index or how closely the electrical and mechanical portions of the transducer are coupled. The higher this coupling is, the less important does the precise tuning or a system become and the higher the resulting transduction e.f.fi~ cienc.y. Losses associated with conversion ot the stored mechanical energy to the "delivered" vibratory energy, however, are not included in the elec­tromechanical coupling coefficient. The power-handling capacity, based on data from various sources, refers to various candidate units under continuous­duty operation with only moderate cooling. The thermal characteristics are of importance because they set limiting conditions on the factors mentioned. The Curie temperature cannot be exceeded without temporary (for magnetostric­tive) or permanent {for electrostrictive) damage to the transducer.

B. Coupler Materials (3o, 59-67)

Parallel with evaluation of transducer materials over the past years, continuing studies o.r candidate coupler materials have been carried out, and the problems involved in the seieetion of coupler materials for specific applications have been recognized.

Couplers for ultrasonic welding syatems ordinarily do not have difficult high temperature restrictions, so the requirements are fairly straightforward. Primarily, the coupler must be made of a material which will transmit high cyclic elastic .forces with low energy losses in the frequency range of interest. In addition, the material must have engi­neering practicability, that is, must be capable of sustaining the static loads imposed on it, must be readily available in suitable sizes, shonld be easily .fabricated, and almost certainly must be metallurgically joinable {weldable or brazable). J

The background, which serves as the basis of this survey has in­cluded studies, sometimes cursory and sometimes in depth, or such materials

23

Table 10

PROPERTIES CJF 'l'R!NSDUCER MATERIALS

Electrostrictive Magnetostrietive Lead Titanate Lead

Mtilti- Zirconate Barium Met ani- Nickel 2V Per-plier* PZT=li · m-5 Titanate obate UAii .r2tm' _. mendur Alfeno1

Electromechanical Coupling 0.674** 0.675 0..50H* 0 .. 40 0.31 - 0.5'1 0.27 Oe27 Coefficient (t33} 0.,30-.35 oso-.60 0.23-.30 0.27-.29

Reported Power- Hanir~ lo4 15 12 8 9 12 Capacity, watts/m --

Piezoelectric Strain-Con- 10-12 256 320 150 90 stant, m/volt

Magnetostrictive Stress 106 16.7-20 32 6.7 Constant (~} i newtons/weber

Curie Temperaturej c•c) 340 340 120 500 360 410 525 500 Density (p}, kg/m3 - · 103 7..5 7o5 5.5 5 .. 9 8.89 8.9 8.15 6o7

1\) Velocity of Sound (c}j m/sec 3960 3780 5680 3125 4780 4790 5260 4500 ,J::"'

Characteristic Specific Imped= 107 ance (Zl?, kg/m~sec

2,97 2.83 2.75 1.84 4,36 4.30 4.30 3,02

Specific Heat~ kcal/kg~•c 0.,10 0,10 Ool2 Ool3 0.13

Thermal Conductivity (K} lo=3 0.30 {kcal=m}/m~sec=°C .

0.30 o.6o 14.5 12ol

The~ Diffusivity {DC}, 10=6 o.4o 0.91 12..5 10.5 m /sec - .. ~ _

Linear Coefficient of 10~5 0.22""o40 0.22 1.9 1.33 1.33 0.95 Thermal Expansion {isotropic)i m/(m-•c) ·I

I :J

Driving rm:pedance = = =IntermediateD~- = = Adjusted by controlling number of >~

coil turns Practical Joining ~thods = = ~Adhesives or Mechanical= - = = - - - = = = =Brazing= = - = = = ~

! Multiply-each item of data by multiplier indicated for the property. **l00°Co *H75°Co 6 Determined for continuous operation with only moderate cooling. #I.Drive voltage (E) is in range of 500-1000 volts/rom thickness or ceramic.

' ~ - ~ t

~\

as titanium, aluminum, R Monel~ K Monel, and, recently, aluminum bronze. The practical application is interesting; for example, the replacement of a steel coupler with one made of K Monel in one type of seam-welding unit permitted an increase of up to 2 gages in the thickness of the material that could be effectively welded at a constant electrical energy input, primarily because the coupler material did not seriously attenuate acoustic energy in the operating frequency range of the machine.

Titanium may be an effective coupler material, since it has a very high "Q~ at high strain levels and, accordingly, transmits vibratory energy with relatively little attenuation. In order to bring the problem of coupling materials into perspective, relevant data for coupler materials presently being considered are summarized in Tables 11, 12, and 13.

The path of vibratory energy is fram the transducer through the intermediate coupling elements to the terminal element or welding tip and ultimately to the weld interface. As indicated earlier, the transmission ot this energy is not straightforward, and careful attention to material properties and acoustical design detail is necessary throughout the entire transmission system.

Maximum power transmission can occur only when the impedances of the component elements are properly ma:tched at their junctions, and the components are made of material that transml:t.a vibratory energy with minimum attenuation. Under idealized conditions, no standing waves exist in the coupling system, and, therefore, all parts of the system are subject to the same cyclic strain and maximum power delivery. As stated previously, ideally the impedance at the junctions between the various components of the transducer-coupling system should match, but in practice this can not always be accomplished (as, for example, at the wedge-reed joint in a wedge­reed spot-type welder).

Table 14 shows the percentage of energy transmitted across the interface between the indicated transducer and coupler materials. This is determined for the case of equal areas from the equation:

where T • the percentage of incident energy transmitted across the interface

pt~ • the specific acoustic impedance of one material (p • density, c • thin rod sound velocity)

p2c2 = the specific acoustic impedance of the second material (68).

*"Q"1 is 2~ times the ratio of the total stored energy at resonance to the average energy dissipated per cycle.

Table 11 PHYSICAL AND MECJWIICAI. :m.OPERTIES (!T RO<M TEMPERATURE)

OF CANDm.ATE COUPLER MATERIALS

Stainless Steel Multi- Aluminum Beryllium Steel (Ser- (Carpen- Titanium ply b7* Bronze Copper Inconel i: KMonel ies 300) ter 883) {6Al-4V)

Physical Properties

Density (p), kg/m3 lo3 7.60 8.23 8.;J. 8.46 7.90 7.84 4.43

Linear Coefficient of Ther-lo-5 11&1. Expanaion, m/{m~-Q.) 1.62 1.67 1.37 1.44 1.73 1.10 0.95

1\)

Thermal Conductirtt} {it), a-(kcal-m)/{m2-sec-•c Io-3 14.7 13.7 3.0 4.2 3.8 6.7 1.7

Thermal Diffusivity («.)**, m2/sec _ 10-6 21.5 16.6 3.2 3e9 4.0 7.7 2.9

Mechanical Properties

Young's Modulus {E), lalO newtons/m2 . _ .. 10.7 11.7 21.4 17.3 19.3 20.4 11.4

Shear MO~lus (~) newtons/m 1010 4.0 4.3 8.3 6.6 1.5 7.8 4.3

Ul ttmate Tenslle · 2 Strength, newtons/m 108 6.2-6.6 4.1-5.9 11.5 6.2-7.6 6.2 7.2 9.4

Yield Strength ( 0.2% 108 offset), newtons/m2 3 1.4-6.5 6.2 .2.7-4.1 2.4 4.7 8.9

.. Mul.tiply each item of-data by the multiplier indicated for the property • "'* tJ(. • K/pS, where S • specific heat.

Table 12

Multi- Aluminum Beryllium plybJ* Bronze Copper Inconel I KMonel

Young's Modulus {E), newtons/m'l

Shear M'odul us { p.) newtons/•2 .

Poisson's Ratio Gr') .

Velocity, m/ sec Shear Velocity (c

8)**

Rod Velocity {c1)***

Impedance j kg/ see-m a 10 7 Shear-(Z )H a cftaract~riStic Speci-, .ric Cz,t)u _ ..

10.7

4.0

0.350

2280 3750

ll.7

4.3

0.350

2310 3800

1.90 3.12

21.4

8.3

0.290

3110 5000

2.65 4.25

17.3

6.6

0.320

2760 4480

2.)): 3.79

* Multiply each item of data by the multiplier indicated for the property.

Stainless Steel (Ser­ies 300)

19.3

1.5

0.285

3140 5030

2.)c.8 ' 3.91

** C8

• ~ andl eL • E/p; CJ. represents longitudinaJ. or thin rod velocity •

..... Z8

• a/iiP and zt· • t!EP.

Steel (Carpen­ter 883)

20.4

7.8

0.300

3160 5100

2.1.,7 3.98

Titanium (6Al-4V)

11.4

3100 5076

Table 1.3

MA.CHININ'G AND JOINING CHARACTERISTICS~~'

OF CANDIDATE COUPLER MATERIALS

Coupler Material Machining Welding - Brazing

Aluminum Bronze 1 1 1

Beryllium Copper 1 1 1

Inconel X 1 1 1

K Monel 2 1 2

Stainless Steel (300 Series) 1 1 1

Steel (Carpenter 883) 1 1 1

Titanium (6Al-4V) 2 2

* lt Not difficult, satisfactory 2r Somewhat di.ffieult ..

References

62 30

309 63

65

30'

30

30

**Data concerning the performance of welded joints are not available.

28

Table 14

IMPEDANCE MATCHING BE'l'WE&N CANDIDATE COUPLER

AND TRANSDOOER MATERIALS

% Transmission Across Interface Between In-dicated Coupler and Transducer Materials

Transducer Materi~s and Longitudinal Impedance (Zt)! 10 (kg/sec-m22

Lead Titanate Zirconate Barium Nickel

Coupler PZT-4 PZT-5 Titanate ".A" (204) Material Z_L 2o97 2.83 2 .. 75 4.36 4.30

Aluminum Bronze 2.85 99.2 100.0 100.0 9,.6 9,.9

BerylliUIIl Copper 3.12 99.4 99.8 99.7 91.3 91.5

Inc~onel X 4.25 96.8 96.0 95.8 1ooiJo 100.0

It Monel 3.79 96.5 97.9 97.7 99.5 99.6

Stainless steel 3.97 97.9 97.2 97.0 99.8 99.8 (Series 300)

steel (Carpenter 883)

3.98 97.7 97.2 97 .. 0 99.8 99.8

Titanium 2.25 98.1. 98.7 98.8 89.8 90.4 (6!1-hV)

29

Recent theoretical considerations (46) indicate that the power that can be transmitted by any elastic system is defined by the equations

2 p .;. ,m m c. v'IP

where P m • the maximum. power

A • the cross-sectional area o.f the coupler or wave guide

c:r-111

• the maximum allowable stress

E • the elastic (Young's) modulus .for the material o.f which the coupling member is made

p • the density o.f this materiaL

The maximum power that can be delivered by a transducer-coupling system for welding appears to be independent of frequency.~ se, but it does depend upon the meChanical and physical properties of tEe materials of which the system is madeo Here ~m represents the maximum allowable stress and viP represents the dharacteris~ic. ~s ecific impedance for the material. Thus, it appears that the ratio ~m /VEP is a .figure o.f merit for the potential of any material .for use as an acoustic transmitter in high powered applications.

Further theoretical considerations {in Appendix) carried out in part during a previous study (45) compared the strain-energy density asso­ciated with the various vibratory modes which are summarized in Table 15. These data indicate that the ratio of the ma:rlmum strain energy to material density, c.,/P, is another way of expressing a .figure of merit for elastic materials.

Application of Hooke's law and simple algebraic manipulation show that the earlier figure of merit is equivalent to E.../p multiplied by the characteristic impedance o.f the material., Thus, it Is clear that either ratio

can serve as a useful guide in any preliminary screening for candidate materials.

The mechanism by which energy is dissipated in the metal coupling members is usually termed internal triction (69). For our application it is desirable that the coupler material offer minimum internal .friction to the transmission of vibratory energy in the frequency range of interest.

30

\,.) H

MUlti-

Table lS

TYPrCAL P8YSIC.AL AND M:.I.!X;aAN.ICAL PROPERTIES OF .C!HDI:Jl.ATE T~ .MttERIALS

steel Inc <mel pliel* M-2 T.;.2 .li3~o X K Monel

Density (p), lb/1n3 Linear Coefticient ot 10-6

Themal kpansion, in./{ in.-.,.)

'lhermal Conductivity (K), (Btu-in.)/tt2-hr-•r_. _

The:nul Dittu.sirlt,- · tc), f't2/br

Specific H~t-'(s), Btu/(ib---p)

Young 1 • Modulus , psi 106

Tensile Strength, psi Yield' Strength (0.2%

ot:f'set), psi

Poissons·Ratio Rockwell C Hardness

Range

0.293

62-66

, ..

Pb7sic8i Properties o.:n2 0.280 0.298

6.2 7.6

as

0.131

o.llS O.lOS

M'9chanioal. Prop8rtuls

31 2$.1 191,000 162,000 180,000 92,boo

0.290 61-66 41 2D-28

*Multiply each item of data by the indicated mUltiplier.

0.3o4 e.o

122

0.152

0.127

31.6 140,000 100,000

0.320 21-28

•

Hol;,b-Rene 41 Ast.rol07 denum. Mo-0.5Ti

0.296 0.287 0.369 0.368 6.S 2.7 3.1

63 936 936

0.095 1.94 2.01

0.108 O~o63 o.o61

46 46 160,000 194,000 102,200 132,000 120,000 142,000 78,800 99,500

0.310 . 0.310 0.310 ,·,

In summary, coupling components shouldt

1. be easily fabricated using standard maChine tools

2. be easily joined; metallurgical attachment is most desirable, and for very high power properly matched systems, it is probably mandatory

3. have good fatigue lite

4. be compatible in characteristic specific impedance to the trans­ducer and te~inal elements; that is, its characteristic impedance must not be too different from the impedance of the other components 11. · ,. 5. exhibit low internal friction at high strain levels and, therefor~, deliver energy with min~ attenuation.

The available information in these categories for materials now being con­sidered for couplers is included in these summary tables.

C. Tip Materials (22, 30, 31, 34, 64, 67, 70)

During the delivery of vibratory energy to the weldment, the terminal tip of the sonotrode is subjected to high dynamic stresses and elevated temperatures for short time periods -- these conditions can quickly damage a tip. The relationship of tip performance to the dynamic stress distribution associated with the tip-weldment interface (46) and to the physical characteristics of various tip materials has been considered previously.

Ordinary tool steels provide satisfactory performance and life in welding aluminum or copper alloys, while Inconel I is satisfactory for welding mild steels, titanium, zirconium, and similar alloys. In welding high-strength, high-temperature, and hard, brittle metais and alloys, the life of the tool steel tips so far used have been short; Inconel: X in the heat-treated and aged eondi tion provides a substantial improvement. Type 301 stainless steel can be welded with a wide range of tip materials, while AM-355 steel is more critical. Evaluation of several tip materials showed only Inconel I to have a reasonable tip lite in welding this material.

The relatively new nickel alloy, AstroloY*, with superior high­temperature properties, exhibited extended life and good welding charac­teristics ill joining several high-strength, high-temperature alloys.

Several kinds of spot-type welding tips have been investigated (Section V of this report). Examples are a .full tip, silver-brazed to the coupler, and a tapered insert tip which is used for certain materials that are obtainable inj'·only l'Od configuration, cannot be readily brazed, or are too brittle .for unsupported use. Previous evaluation studies have included

* Product of the General Electric Company.

'32

full tips of tool steel, Inconel X, molybdenum-0.5 titanium, Nitroloy, and other metals and have inCluded inserts of tool steel, tungsten carbide, titanium carbide, X Monel, and austenitic manganese steel. Some of these materials were found to crack under high loads, some eroded readily and required frequent redressing, and some e:xhibited excessive sticking to the weldment.

Information, regarding the various welder-tip designs, is ~ marized in Tables 17 and 18 in Section V. The tip material must be tough and resistant to wear so that the tip does not deform, spall, erode, or crack when high vibratory power is applied; also, satisfactory physical properties must be retained at elevated temperatures which depend upon the materials being welded. Tip materials with good thermal conductivity are desirable because liquid cooling of spot-type welding-machine tips has already become a standard machine feature.

However, in the final analysis, tip materials must be tested under actual welding conditions before a proper evaluation can be made • .Accordingly, performance data for the more promising tip materials will be obtained and reported as this program proceeds.

Information concerning the physical and mechanical properties of promising materials for the fabrication of sonotrode tips and disks, is summarized in Table 1$. At the present time, this information is incomplete, but additional information is expected from replies to our letters ot inquiry.

,-,>! .(

33

IV. ACOUSTICAL MA'l'ERIAI.S STUDY

Determine the material or combination or materials tor. the trans­ducer and associate equipment most efficient in producing a dis­tortion-free solid-state bond.

A. Transducers

The suitability or candidate transducer materials tor use in ultra­sonic welding equipment is ~ing evaluated on a basis that reasonably appro:x:i­mates .the end use. Witbspot-t;ype welding equipl!J8nt~ the transducer receives a pulSe or electri.cal energy or4inarily or less than 1-second. duration. In roller-se8.lll-typ& equipment the energy is applied continuously. In spot-type welding the transducer mar ha..-e a relatively easy duty cycle in which its thermal inertia parmi ts the acceptance or high power levels. In continuOllB­seam welding~ steady-state conditions are likely to prevail so that trans­ducer cooling exercises a strong influence on the energy which the system Will handle.

. .. The energ7 requirements for welding the refractory materials of interest in this investigation were estilllated on the basis of the energy equation (discussed more tully in Section II) and by extrapolation from pre­'Yi()118ly obtai~d da'tia for these or similar ma.terialS. These en&l"IJ' require­ments will be refined from time to time on the basis of e:xperimental data obtained, as the work proceeds.

Tran$d'l,lcers are routinely evaluated ( 71) by obtaining a motional impedance loop, the data of which dat'ines the resonant frequency of the transducer, . the "Q" of the transducer and the potential transducer efficiency. Such loop data are orciinarily secured by means or impedance bridges equipped vi th oscillators and detectors. The motional-impedance loop 1 however 1 is general.l;r ascertained at/ instrument power le'V8ls. Previous experience (lJZ) has shown that transducers can be evaluated for purposes or interest by a direct calorimetric technique which can provide important ancillary inform­ation. The transducer is attached to a coupling member in a manner essen­tially the S&DI8 as it is attached to a welding machine. The coupling member, is connected directly into an energy absorber such as a large block or lead (Fig. 3) 1 in which a cooling coil carries away the vibratory energy which is degraded to heat in the lead billet.

It ~11 be appreciated that the calor~etric method permits «4'ivinc the transducers at elevated power levels vnich can be either pulsed power ..

34

Iapedance Matching Section

Inter:.diate hr17 Coupler

Water Outlet

Imnllation

VibZ'atOJ.7 ~81 Absorbing Radium

{Lead)

Fig. 3: CALORIKETRIC TEST SCHEME FOR TRANSDUCER AND COUPLER EVAWA.TION

.t. '.

required for spot-type welding systems or continuous power as is necessary for continuous-roller-seam welding equipment.. Moreover, problems of trans­ducer cooling can be studied; in the interval covered by this report, the evaluation system shown in Fig. 3 was designed am partly fabricated. It is expected that this system will be in use to evaluate transducer materials prior to the end of September. Information will be developed on the better­known camidate materials first, particularly lead drconate titanate, the electrostrictiTe PZT-4 from Brush Development Company, and the magneto­strictive nickel cobalt alloy from International Nickel Company. Basic information on standard "A" nickel will be obtained. Information on the over-all efficiency of these materials under p~ered conditions which approximate end use in ultrasonic welding should be shortly available.

B. Couplers

As described in Section III,.,a coupling member which intervenes be­tween the transducer am the point of energy delivery into the weld.m.ent pre­sents three distinct problems:

First: It involves energy losses due to reflections at the transducer­coupler interface and at other junctions or discontinuities between the transducer am the point of' energy delivery. Reflection losses at such junctions are minimized when the acoustical properties of each material are about the same, as reported in Table 11..

Secorxl: Losses occur in a coupling member as a result of elastic hys­teresis (that is, internal friction); such losses are affected by both power level and frequency, with greater losses occurring at high power leTels ani frequencies ..

Third: If the coupling member conducts vibratory energy at a suffi­ciently high-power level, or if the standing wave ratio in the system is large and a high-power level is involved, ordinary fatigue failure can (and, indeed, does) occur.. When the standing-wave ratio approaches unity, ho11ever, a low-carbon steel bar conducts vibratory energy com­parable to the electrical eneru carried in a copper conductor (upwards or 10,000-12,000 watts/cm2). This energy range can drop as law as 100 wattjcm2, however, when the standing-wave ratio is high. Therefore, it is exceed~"lgly important that a coupling system be designed to mini­mize reflection and transmission losses and to utilize materials with outstandingly good mechanical properties in order to maximize the energy delivered to the weldment. All these factors are difficult to obtain in a single material; in fact, little is known about the trans­mission and the fatigue properties in the frequency range and at the vibratory energy level that is implicit in ultrasonic welding equip­ment design.

As reported b;y Mason (48) such information can be obtained by carrying out certain measurements involving acoustical velocity transformer elements. More recently, Neppiras (72) developed a technique for studying

36

materials of interest in high-power ultrasonic transmission investigation. With the Neppiras method 1 both internal energy losse~;~ and fatigue strength can be determined.

With the standing-wave-ratio method ot energy measurements 1 which adds an important factor ot control to the Neppiras evaluation technique, reasonabl7 straightforward detel'lllinations can be made under reproducible conditions in a relatinl.7 short period or time ..

During the interval cOTered. by' this report, equipment has been devised in accordance with Neppiras 1 technique, and partially' assembled.

c. !te

1. Spot-!m 'rips

Spot-type tips are used to produce repetitift welds in a giftn material 01" combination at materials o In this system, the sonotrode tip sel"fts two purposes 1 the aost itlportant ot which concel'!lS deliver;r ot the eneru to the lf'8ld zone. Second]J' 1 the tip provides a terminus to the acoustic system. For reasons discussed prertousl.7, the material selected tor the sonotrode tip must be tough, resistant to wear and match the impe­dance ot the weldllent as closel.7 as possible.

'fhe effect ot coupling between the tip and the wel.d:ment, as influenced b7 materials 1 frictional characteristics 1 etc., has not been i.nve'stigated. A 'V'8.riet7 or materials have been welded, hawe'ftr, with big differences between the physical and mechanical properties of the tip and wel.d:ment •teri&ls. For instance 1 alumillWil alloys and low-carbon and stain­less steels 1 as well as other engineering materials 1 have been welded success­ful]J' with toc;>l steel and nickel alloy- sonotrod.e tips o Furthermore 1 coupling can be frictional, as in spot-type welding or sheet material, or by' positive driV&, as in the welding ot joint geometries that permit mechanical locking between the tip and the weld:men.t... A. high modulus, toughness, and resistance to wear are required properties for resistance (frictional) coupling tips 1 while in the positive-drive arrangement, the major requirement is adequate mechanical strength to support the expected static and dynamic stresses.

2o Continuous-Seam or Roller Tips

'fhe basic requirements ot roller or continuous-seam. tips are alm.ost the .same as those ot the spot-t7Pe tips. Roller tips, however, are ot two types, ~sonant and nonresonant. The nonresonant-type tip serV&s as in the spot units, as a terminus to the ac011stic system and is subjectJ to oil.Q' the d.ynamic shea~ and normal stresses attemant to the ener17 deliver;r into the waldJMnt. 'fhe resonant tip serves a dual role or delivering the energy to the wldllent and or functioning as a· ~~&tching trall8f ormer between the acoustic system. am the wald.ment o Thus 1 the rollers m&7 be subjected to the dynamic vibratory stresses associated with component resonance and to those stresses at the junction to the vel.dlRent associated with energy deli very to the wald.

37

The lite ot roller tips is evaluated in terms or producing seam welds of uniform. quality. Tip lite can be c~nsiderably improved by geo­metric considerations insofar as these govern the surface fibre stress for a given disk deflection. Iuasmuch as the power delivered is proportional to the B~ or the displacemebt1 and inasmuch as the disk fibre stresses vary as the cube of the diSk deflection with respect to the neutral plane, material properties become even more important. Disk materials must possess high strength, extended fatigue lite, and compatible coupling and impedance characteristics.

The .criticality of t-hese components 1 as demonstrated during the evolution or the welding process, has been indicated. The suitability o.t' the various materials .t'or power-handling capacitT, vibratory energ;r trans­:ad.ssion characteristics 1 impedance -.tching characteristics 1 and fabrication and joining adaptability can be ascertained by measurements and tests o No simple •asurement_, however, iB immediately an.ilable for screening purposes, and reliance lllUSt be placed, at least temporarily, on staming-wave-ratio 118asurementll to naluate energ;r delivery by any unit at constant input conditions arxl on weld qualitT• Obviously, this iB quite a laborious pro­cess, arxl only a limited liU11lber or candidate tip materials can be studied.

38

• Qj

V. ENJmGY DELIVERY ME'l'Hoi>s

Determine the most efficient methods ot supplying vibrator7 energ7 to the bond interrace.

•• !Z!tems

A variet7 or ultrasonic -.lding arrays tor spot-type, rolle~seD, and ring-welding a;ylltells have been developed, but details ot the specific advantages and disadvantages remain to be defined.

In gener~., there are two broad classes. ot s;rstems which are in­dependent ot the wld geometey. The .first embracee all those types in whi~h a reaction element, t:1r anVil, supports the work pieces and resists compliance

·thereof with the vibrator,. forces exerted b7 the powered sonatrode. 'l'he second or •opposition-drive• class ccaprl.ses systems wherein vibrator,. energ i81 &!livered to both sides or members ot the weldm.ent -- no massive reactive element such as an anTil is involved.

The class ot welders havfn& the reaction eleme.nt includes the wedi~-l"eed design in which the re~c;l is excited by a single Qoupler element {wedge-1;ype) or .b7 two diametricall7 opposed .couplers operatins in oppo-s. it.io. n {Fig .• U of P• 3), thus ettectivel7 doubl. ing~ the .. power capacitJ• ~~. cat~gor,. also includes the lateral-drive coupler system (Fig. lB), the roller system (Fig. lC), and the torsional system (Fig. lD).

~ecise relative efficiencies ot the wedge-reed, the lateral drive, and the ring-welding systems have not been established, although much data have been obtained previously with all three types. In general, the wedge-reed 97~tem has been used vi th higher power equipment and the lateral-drive with lower power arrays. '!'he higher bending loads which are associated with the application ot clamping force to heavier, harder ma­terials limits th.e standard lateral-drive system because ot such. second­order ettects as •tip bounce• unless special provisions are provided to preclude them.

'l'o consider. the potential etfiC:l.elfCY ot various types ot spot­wel~r. S7stems, a. theoretical anal;rsia relating the. strain ~nerg7 d.ensit7 to amplitude tor .the longitudinal. and_ fl.~~ ,case.s, ,._viousl7 Carr'ied out, was. extended to include the torsional concept (in .Jppendi:x:). This more ccaplete analJBia indicates that torsional. and longitudinal modes are comparable in powero-handling capacit7 ('!'able 16), whereas the lateral or bending mode involves greater stresses at similar amplitudes.

39

fable 16

RELATIONSHIP BETWI!'.EN RELATIVE S'rRAIN ENERGY LEvELS

PCB CONSTANT J.MPLITUDE AND RELATIVE .AMPLITllDE

F(Jt CONSTANT STRAIN ENERGY LEvELS

Constant Amplitude1 Constant Strain Energy Mode of Relative Strain Densit:y1 Relative

Vibration Energy Density Amplitude at End

:t.ongitudinal 1.0 1.0

:Lateral• Round 2.4 0.65 Rectangular 1.8 0.75

Torsional 1.0 1.0

nemra1 {disk)

5.1 o.45

# K ~& m va

in e~isting welders of significant power-handling capacity, the reaction e!~t or anvil is a pote~tial source of energy loss because ot the e~rgr Vhich passes t~()ugh and beyond the weld zone. . To minimize these losses, ccmsiderable effort has been devoted to anvil developm.ent. As a result ot ·this p!J.f!lt 1fork., isolation systems for anvils have been developed and are now in da;y-to-day use.

On thebasis of the information of Table 5, it is anticipated. that acoustical power up :to about 25 kw, and possibly higher, will be required to join. the candidate materials in thicknesses up to 0.1 inch, and that power levels of this order will almost necessarily be delivered via both of the clamping sonotrodes.

The opposition-drive class of system does indeed, eliminate the necessity for a massive, noncompliant anvil and the problem. that are en­tailed. Such systems have been developed and utilized successfully. How­ever 1 unless the design ot an opposition-drive system ipcorporates solutions to problems peculiar thereto, it is possible that the energy losses will be greater than those experienced with the reaction-sheet type. For example, a slight shifting of phase in the tip excursion of either sonotrode (from the lBo• out-of-phase condition that must prevail) will abruptly produce a great d,ecrease of energy deliveq. As a matter ot fact, l.Ulder certain cir­cumstances one transducer coupling system may act as an alternator with the opposing system acting as a motor so that almost no work· will be done at the weld locale.

There are at least three avenues to satisfactory opposition-drive operation Which have previouSly been investigated and developedt

1. . mech&nical intercouplirlg in ,.ih~ch all the transqucers. drive . a cCBI.on coup].e:r, and ·the energy ·output of the coupler is di v1decl by means ot a locked mechanical out-at-phase system to pravide lBo• out-of-phase displacement to the sonotrode tip

2. e#tct~ical intercoupling which involves standing-wave-ratio or other monitoring equipment on each coupler or tip for detecting and automatically maintaining the proper phase relationship by a servotechnique

3. electromechanical intercoupling whiCh utilizes a combination of these techniques.

···sa- exPer:imentation is .still required to select the best of these two classes and also to evaluate the practicality of the types ot transducer­coupling systems tor use in heavy-duty equipment.

Ji &_!f&!tt

In previous studies on this problem, available equipm.ent vas jury­rigged to operate at relatively high powers (up to about 6ooo-8000 watts of high-frequency power to the transducer) without regard to the practicability of u1 timate use in heavy-cmty equipm.ent. A refined approach is now being prepared to obtain additional information on the opposition-drive class of system.

The possibility that ring .welds may be more. desirable than standard ·spot-type velds is also being considered. In any spot-type veld, the struc­tural load is carried from one side of the veldment to the other, generally through the periphery of the. spot. The center of the spot contributes little to the spot strength. Nevertheless, with the ordinary ultrasonic spot-type welder, energy is used to produce the interfacial disturbance over the entire weld area, including the center. A ring weld has the promising stru.ctural advantage of a bond generated only Where it is useful, at the periphery. The ring configuration can be adjusted to provide not only a large bonded area, but also a large-diameter, more efficient weld.

B. Components

1. Transducers

On the basis of information compiled to date, it appears .that trans.ducers for welding will involve either magnetostrictive laminated­sheet metal stacks or such ceramic-type materials as lead zirconate t.ita­nate. Much experience has been accumulated with the des;lgn, fabrication, and. service life of SUCfh magnetostrictive materials as "A" nickel and nickel-cobalt alloy. ~o far as is known, little, if any, experience has been obtained vi th high-power ceramic transducers capable of . sustained energy delivery via metal couplers. In order to obtain practical veri­fication of the reported theoretical performance or such ceramic ma­terials in large transducers for extended operation, certain designs for such transducers, evolved prior to this work, have been partially eval­uated {Fig. 4).

Loading to maintain the ceramic elements in a state of compression is variousl7 applied. In Fig. hA., peripherally located tie bolts produce this c011pressive loading via end plates. In Fig. 4B, a center tie bolt serves the same purpose, vhUe in Fig. 40 and D (assembled and exploded), the containing tube carries the tension reaction.

It has been determined that the design of Fig. 4A exhibits spuri­ous plate-type resonance, but this unsatisfactor,y condition may be elimi­nated when one plate is bonded to a metal coupling bar. The design of Fig. 4B has been difficult to evaluate due to a lack of symmetry between the inner tension bolt length and the outer (slugs and ceramic washers)

42

A: PERIPHERAL 'l'ENSI ON BOLTS B: CEN'lER TENSION :OOL'l'

C: .ASSEMBlED TEHSION SHELl..

D: TEISIOH SHELL mSASSEMBIED

Fig. lu TRABSDU'CER DESIGNS IICORPCBATIBG CERAMICS

&&£a ' &X.t&,,,& . I

compression path •. Effort will be made to correct these inadequacies as the work proceeds. The design of Fig. 4C and hn, in which the tension reaction is carrie~ in the enclosing tube, has just been completed and has not been powered for evaluation. These transducers incorporate lead zirconate tita­nate (obtained frODt Clevite) into the preloaded type of mechanical assembly which precludes the need of adhesives, offers good probability of satisfac­tory cooling, and, especially avoids cyclic tension loading of the ceramic elements.

Transducer efficiency will be evaluatedby calorimetric investi­gation with the system of Fig. 3, as discussed in Section IV.

2. Coupling Members

An efficient transducer, while very important, does not insure power delivery to the work pieces. High energy losses can occur in the coupling system between the transducer and the work, especially at high power delivery. Vibratory energy is converted to heat within the coupling system by internal friction. For small deformations (low power) the loss per cycle is low because essentially perfect elastic behavior prevails; at stress levels associated with high power delivery, the problem of in­ternal friction losses is serious.

To our knowledge there is at present no satisfactory theory for internal friction in solids that embraces a broad vibratory frequency spec­trum, although such losses can be measured by several experimental methods. For example, at low stress levels (on the assumption of simple harmonic motion) the natural logarithm of the ratio between successive oscillations {log decrement), as determined with a torsional pendulum, maybe used to estimate the internal friction losses •

. Many investigators (73-7~) have worked at frequencies up to about 200 cps, and same work (76) has been conducted at high frequencies. Except for the work by Neppiras (72), little information has been located on the losses in various metallic materials at frequencies in the range of interest, 5-50 kilocycles per second.

Letters were dispatched to both foreign and American individuals and organizations, who, we believe, can supply infomation on the internal friction losses and acoustic transmissivity of materials in the frequency range of interest.

Parts for a test system designed on the basis of work by Neppiras (72) have been fabricated and await final assembly. This array will be utilized to determine relative, and possibly absolute, acoustic transnis­sivity of candidate coupler material,. Test specimens of the· immediate candidate alloys (tool steel, K:onel, and al umimmt bronze) have been de­signed and fabricated.

. 44 ~.- '·-

3. Spot-Type Welder Tips

The problem of attaching welding tips to the sonotrode and/or anvil cannot be ignored; mechanical attachment, while feasible at modest powers, has not been as reliable for higher levels of power; brazing attach­ment of tips is known to be practical at high power levels. Thus, at least for the present, dne to independent considerations, tip materials Should be brazable if. this is possible.

Information, regarding the various designs of spot-type-welder tips is summarized in Table 17. Mechanically attached tips are highly de­sirable if not absolutely mandatory. Examples of mechanically attached tips are Types 3 and 6 of Table 17. Type 6 is the more desirable, for a variety of reasons, but, especially, because it can be fabricated easily from small pieces of material (often necessary when a new or special alloy is involved}. Type 3, however, is difficult to manufacture and, conse­quently, is more expensive because a modest quantity of tip material in a variety of shapes is frequently difficult and costly to obtain.

4. Roller-Seam-welding Disk Tips

While spot-type welding tips constitute such a small part of the welding system that their acoustic properties can be neglected, disk tips for roller-seam welders are a critical factor in resonant systems since they must . transmit vibratory energy fran the center to a point on the periphery. Disks tor roller-seam welding machines are sophisticated, and their design has been the subject of various theoretical treatments and experimental measurements from time to time. Since such disks continually place fresh cool area in contact with the workpiece, they may not involve as rigorous metallurgical and physical demands as spot-type1elder tips. These designs have definite boundary acoustic conditions, howeTer, and be­cause of stress buildup in the center of the disk cannot be indefinitely extrapolated to higher powers. Hysteresis can cause energy to be absorbed within the disk; unstable operation and an unusual type of metallurgical failure may result.

Information concerning various roller-seam-welding tips is sum­marized in Table 18. The Type 1 tip is an operable nonresonant mass, but any reasonably high welding rate involves an unsatisfactorily high angular velocity of the transducer-coupling system. Types 2, 3, and 4 are charac­teristic resonant disks, showing several disk-to-coupler attaenment methods. Type 5 is a resonant toroid that has received considerable attention.

45

I'

!!£!

1

2

~·

s

6

GeoJHtr,y

-

-

'1'able1?

OHWCTHRISTIC:S or 'l'ERMDIAL TIP-REED ARHAIS PUR SINGLE-SPOl' WELDERS (Orientation for special applioationa ~aaible with oontoared ttpa for all trpea.)

Eaae of Weld. iDe Duoription A\taoblent Fabrioatioa Rep~o-t &ting JunotiDn Performance

Spherioal. or Bras eel RelatiftlT r•-oonaai:lc Sat ilfaotor,y aoul.pt.uNd .. .,. at 1olr power work aurtaoe le'Nla

Oont.our.d for Bra sed. Some~~ hat '1'--00DI'tllli:lc Mile\ be pre- S&tiataotor,y greater •tiDe d.itficul~ oile. IUl:'faoe Brasing re-

quiNa akill.

Meohan1cal.l7 'l'bread.ld

Miaoellaneowt

Joint 1a eon in 8Mart eapeci.allT wben 'Rl'll1

aDd. it. qual.it,' 11 d.syi-Olll.t k iDIIU'tl .

Brased. jotat 1a not bigbJ7 loMed. in ahee.r

:.+---'--'-. attaobecl t~

-Muitiat.ep \~ Bra sed. Ditticml.t '1'-.ooDIN!IIing

-S!nc18-at.41p Bra sed. Relat.inlT '1'--00QBUIIiDc .. .,

-Preee-fit.tad Ditf1oult.

-

Mwtt be pre-aile

T~ M8ilT looated. for bra sing

S&t.iltacto17 Suo~ in reduciDc ebiiU lo.d on joint tor hJ&b-ponr appli-oat. ion

S&t.iafaat.oJ7 B.tlcluoed. lbiiU load OD joint

Satiafactor,y Inalll't. _,. be of d.itfi­oul t.-to-Mohine ..tal. Preea-ttt. or high power 1111117 d.efft'll thin end. of reed

'l)pe

l

~

Table 1.8

CIIAilACTBRISTICS OF TBRMDAL TIP.COUPLIR ARRAYS POll COI'l'~ ROU.ER WBLDIRS ('l'J'pu 2 to 9 haw Nd.uoecl :rotat.ioD&l lpMCl 11114 illp:roftd work ol..,_,. :l.a o~ to fJpe 1)

Velooit7 Tru~~to:r-•tin

o.a..tJ7 Tip Coupler Jomt. Ratio

==J JlomoHo- 1J:poMrlt;.1al Tbrea41Cl 1.0 _, ...

0.1

0.1

... ot Pabrioa t:l.oll

BelaU'ftlT .. .,.

Ditt1calt arr4/or apml:l.ft

DNip tor IllpediiDOe lfatoMDc

CaD be ad-jutiCl

!lore d.Uti­OIIlt t.bll.ll tor 'l)pe 1

Jlore dUti­nlt !;.ball tor fJpe 1

Jlore dUti­nlt !;.ball tor fJpe 1

JUaoea.liMOU

• .,. w np1aoe

li&llatlrU ... at ,Juat.oaue tailan !all' to replace

Higbl.J' et:rHIICl center •dilk area el4!Wiat.ec! Serio'QI •io:ro­ld.l:l.aU.O . p:robl- at jUDCUOIIl oaue failure.

11dclflll ~

Oooll, bat IIOt ~ O'IW liiiPb ot ...

Colllliltat Oft1' ltllllltb ot ... , S&tiltaotol'T veld qaal.ttJ'

Coal 18tat Oft1' l.actb ots._ PGver del:l.ftl'T ~to that of dilkl lip to 2000 v S&tiltaotol'T veld qul.:l.tJ'

~

APPDDIX

'1'HB LIMITATial ON AMPLITUDE SET BY MAIIMUM

S'!'RUlf BIERGY Dr VIlit.l'l'llfG SYSTEMS

In JI8Jl7 applications ot ultrasonics it is desirable to aChieve as

great an aplitude ot oscillation at the work area as is permitted b;r the

elastic propert18s ot the materials constituting the vibrating system. It

is aSSUDied in this analysis that a given isotropic material is characterized

by a maximtmt pel'lliss:i.ble oscillating eiastic strain enera densi t;r t which

can not be exceeded. without fatigue failure, regardless ot whether the enera

density' is associated with shear distortion, simple compression, er a cembi­

nation ot the tvo. !he treatment can be moditied later, it it turns out

that the fatigue limit depends on the nature or the elastic distortion.

Longliud.iluil. 'fibration ot a 'Oilitorm Bar

Consider tirst the longitudinal vibration ot a Blender halt-wave

rod ot unitol"'ll section. The strain at 11D7 position X, with origin at ihe

center ot the rod, 18

g · (~~). cos r x.

where X bas the range - ~/4 S X :S ),14,

... 2• '·' r-"'·r,

48

(1)

(2)

'.rtie maximum amplitude at the end of the rod is )l

'.. . ~J.14 cos ltx di • t ~- .

'the ma:d.lllUJil elastic energy d.ensit7 at the center is 2

c. • ix ~~). where E is Young's Modulus;

hence,

s. - i E x2 '-2- • ~ P w 2 ,._z.

(3)

(4)

(S)

Since the marlmwn velocity- at the end ot the rod is (.0'. • ~.~ lq. 5 can

be written

e:. -i p 5!, {6)

which is the kinetic energy per unit vollDIIB of the material at the end of

the rod. Whereas the kinetic energy densi t;y and ~oci t;y is independent

ot trequeno;y tor a given upper li11lit to e.., the peraissiole amplitude

varies inversel;y wiith .f'requenc7.

Lateral Vibration of a UDitor.a Bar

lext the hee-tree lateral vibration of a bar 'of circular section.

!he following results from Rayleigh, p. 281 et seq., (77} can be used. For

the frequenc;y,

(7)

49

For tbe maplitude at the end, in tel"Dl8 ot the amplitude at the center,

(8)

From the table on p. 282 ot Rayleigh (77)1 by taking secon.d ditterences,

~ 2~ - 29.1 . • 11.1 -4 Al:i. 'center T .lend • ~X " .

(9)

The lll&X1mtll'l tiber strain at the center is

. 2

(-H).- & ~\ - 17.7 it .8114 t '() xJ center (10)

On coabiDiD& (7) and (IO}

(11)

Bence, troa Bq. (4)

c - . rr_. 2: ·2 l c;;. 2.37 11 p cJ) . 'en.!J • (12)

This result shows that tor a gi"t'en amplitude at the ead, the (surface)

strain-energy densi t7 at the center is nearl7 two and. one4uu..r tiaes as

great aa tor the longitudinal case. It depends on clens:l, t7 and .trequenc7

as betore, a rei!S1il.t that is obYioua t::roa dhlensional considerations.

I1 the 'bar is ot reataaplar section, ~ tb.iclnesa 2a, Bq. (1) " ,·~

bee-•

.., ·: v.l.. C~~.m' Tt off • (13) J

since the radiuS of Qr&tion of the section is DOll' a/{'! iastead ~ a/2 • •

so

!Jeftoe, ,~ value of a/.,!2 in :Eq. (10) is decreased·:,by the factor IJ/2 (for

2 a 8~~ fre~eney) and the energy density of Eq. (12) by ( ~/2} • 0.75.

4eao,diJ4Bl7, a rod ot rectangular section is superior to one of circular

•~nt..-, ~,_ as large an amplitude ot vibration as possible is desired.

4lllt, r'm!Jcm ot a tld:a Uhitora D~llk

It can be showa (78) tor one nodal circle with tr • 1/3

• !II 2.615 , . . 2! ' t fi5> a p(l-d"' )

(14)

~ a 18 the radius and t, the thickness or the disk. The shape of

the cU,Jt ~ aiven by the tanction

W-. J0

(kr) • .:>\Ic,(tr) J

' _.lt 2, 4 w5,-., r " ,., , ((_ , Ia s t 2

~· • ~ ~(i-r1)

Jl. (ka} A "• fi (b)

(15)

........ ~IMt J aDd I frmction are ordina.J:T and aOditied Bessel f'cmctioll8,

~.,_fi!..Xy. For II"' • 1/3, ica • ).Oi and ..>\ • -o.oelci. The aaplitude

'' ~ --.. l• 0.74 thai at the center.

j calcaiation based oa Eq. (.15} shows that the curvature at the

.., .. , tO'l' a diaplaC8Jilent &~~plitude .j.center' is

(16)

•

'!'he strain at the surtaoe is, therefore,

f-ti,) • j 4 • i ~ (3.01)2 1

•; b~:]f8 • 3.62 -y8t Aedge' {17)

\," Y 11 J:r a •

o:a introduci:ag the numerical values alread7 quoted for ka, )\ and

l. · · 11 ...... /1. t The strain energy density for a plate stretched Ull.:if'oraly · e"'6.. cen er.

in all directions an aaount ii/ai is

e ·r!r (H)2· (18}

On introducing the value frOifl Eq. (17), for ?J J/ax, the frequency from

Eq. (14), and r • 1/3, 12

e •• f'l (3.62)2 p(l; J l , 0)2 12 • r! 1·0 ftl 2 •. 2 J (19} • edge ;}• ' pA> .a. edge •

{2.615)2

Bence, for a given 811Plit1lde at the eqe, the JU:rl:aum strain energy density

is slightly aore than five time& that of the longitudinal case for the Salle

aapli tude and trequenc7.

Torsional Vibrations of a tlliitora Rod

Consider, finally, the tord.onai dbra~ions ot a li.'Difol"Jit half-waft

rod of circular section, vith origill at the center. rt 0 is the arigular

d.isplaoemeD.t at an7 section, the •gu.lar strain is

-H- (J~. cos ( x, (20)

(21)

52

,e ss as usus x , , c. i

'1'he linear am:pli tude at the outer radius a is

{22)

where {;; :f/FJX)JJ. is the ma:x:imma shear strain at the surtace of the rod.

Since the strain energy density is

£. ~ E*) 2• {23)

from Eq. {22) and (23} 1 by 8llbstituting I • lAl/ct and et • ,;;:!;, e. - i p (1.)2 '-2. (24)

The torsional case, ther8f'ore, is identical with the longitudinal

ease discussed in the first section, the longitudinal v.ibration of' a unif'ora

bar, of' the published material. All of the results obtained show that

e../P is a figure of merit tor an elastic material, which can be used. to

estimate the largest possible vibratory amplitude at a given frequency,

regardless of the geometry ot the rlbrator.

S3

fi

LIST OF REFERENCES

1. Jones, J. B., and Je J. Powers, Jr., "Ultrasonic Welding", Welding J. Vol. 36, Aug. 1956, pp. 761-766. .~

2. Jones, J. B., "Progress Report on Ultrasonic Welding", Minutes of the Sixth Annual Meeting of the AEC Welding Committee, Oak Ridge National Laboratory, Oak Ridge National Laboratory, Oak Ridge, Tenno, Sept., 25-26, 1956.

3. Jones, J. B., "Ultrasonic Welding", Fabrication of Molybdenum, Cleveland, Ohio, 1959, pp. 88-102.

4. McCarthy, D. v., v. Pirc, and w. Hannahs, "Ultrasonic Welded Aluminum-. Copper Junctions as Electrical Connections", Reliable Electrical Connections, Third Electronics Industries Association Conference, Engineering Publishers, New York, 1958, pp. 113-119.

5. Anonymous, "Advances in Fabrication Techniques Revealed at Southern Metals Conference", Metal Progress, Vol. 76, Aug. 1959, p .. 128.

6. Jones, J. B., and H. L. McKaig, "Ultrasonic Welding and Improved Structural Efficiency". Presented at the 28th Annual Meeting of the Inst., of Aeronautical Sciences, New York, N.Y., Jan. 25-27, 1960 ..

7. Aluminum Company of America, (Collins, F. R.), "Ultrasonic Welding", Report No. 2-57-5, New Kensington, Pa., Feb. 25, 1957.

8. Kaiser Aluminum & Chemical Sales, Inc., "Ultrasonic Welding", Kaiser Aluminum Sheet and Plate Product Information, Chicago, Ill. Sec. Ed., January 1958, pp. 235-237, 133.

9. Aluminum Company of America, (Collins, F. R G) , "Properties of Aeroprojects Ultrasonic Seam Welds", Report No .. 2-59-14, Process Metal­lurgy Division, Alcoa Research Laboratories, New Kensington, Pa., Aug. 7, 1959. ·

10. Crucible Steel Company of America, "Ultrasonic Welding of Titanium" 1 Titanium Reyi«r, Vol. 8 (March 1950), pp. 5-7. :

11. Fabel, G., "Ultrasonic Welding: Optimizing the Variables", Assembly and Fastener Engineering, VoL 3, Nov. 1960, pp. 32-36.

12. Terrill, J. R., F. R. Collins, and J. D. Dowd, "Applications for Ultrasonic Welding of Aluminum", Paper No, 60~A-322, Presented at Winter Annual Meeting of the ASME, New York, N. Y., Novo 27-Dec. 2, 1960.,

13. Alden, J. H., "Ultrasonic Sealing of Foil", Modern Packaging, Vol. 34, July 1961, pp. 129-133. ...

14. Koziarski, J ., and P. Dick, 11Ultrasonic Welding Joins Stainless to · Aluminum in Nuclear Power Plant", Materials in Design Engineering, Vol. 53, May 1961, pp. 146-147.

15. Battelle Memorial Institute (Weare, Norman Eft, John N. Antonevich, et al.), "Research and Development of Procedures for Joining Similar and Dissimilar Heat-resisting Alloys by Ultrasonic Welding11

, ASTIA Document 208323, WADC Tech. Report 58-479 on Contract AF 33(616)-5342, Project No. 7-(B-7351), Feb. 1959. ·

16. Welding Research Council, "Current Welding Research Problems", Welding Journal, Vol. 37, Aug. 1958, Research Supplement, pp. 379-s to 384-s.

17. Koziarski, J ., "Some Considerations on Design for Fatigue in Welded Aircraft Structures", Welding Journal, VoL 38, June 1959, pp, 565-575.

18. National Research Council, (Materials Advisory Board, National Academy of Sciences), "Summary Report of the Committee on Refractory Metals", Report MAB-154-M, Vol. 1, Washington,, :Q. _c_ .. _, Octo 15, 1959.

19. American Welding Society, "Ultrasonic Welding", Chapter 52 in Welding Handbook, Section Three, Fourth Edition, New York, N. Y ., 1960.

20. Welding Research Council, "Current Welding Research Problems11, Welding

Journal, Vol. 39 (Dec. 1960), Research Supplement, pp. 547-s to 552-s.

21. National Research Council· (Materials Advisory Boa~ National Academy of Sciences), "Joining of Refractory Sheet Metals11 , Report No. MAB-171-M, Washington, D. c., March 20, 1961.

22. Koziarski, J ., "Ultrasonic Welding", Welding Journal, Vol. 4o, April 1961, pp. 349-358.

23 • Jones, J. B., N. Maropis, J. G. Thomas, and D. Bancroft, 11 Phenomeno­logical Considerations in Ultrasonic Welding", Welding Journal, Vol. 40, July 1961, Research Supplement, pp. 289-s to 305-s.

24. Neville, s. W., "Ultrasonic Welding11 , British Welding Journal, Vol. 8, April 1961, pp. 177-187.

25. Kitaigorodskii, Yu. I., M. G. Kogan, et al., "Joining of I1etals in the Solid State by Ultrasonic Vibrations 11 , lzvesti~a Akademii Nauk SSSR, Otdelenie Tekhnicheskikh N~uk, Aug. 1958, pp. 8-90.

26. Ainbinder, s. B., "Certain Problems in Ultrasonic Welding", Svarochnoe Proizvodstvo, Vol. 32, Dec. 1959, pp. 4-6.

27. Silin, L. L., v. A. Kuznetsov, and G. v. Sysolin, "The Ultrasonic Welding of Aluminum and Its Alloys", Svarochnoe Proizvodstvo, Vol. 33, April 1960, pp. 42-44.

55

•. a .. £&&&&_ &.ta.c

28. Garber, R. I., L. M. Polyakov, and G. N. Malik, "Ultrasonic Welding ot Copper" 1 Fizika Metallov i Metalloved.enil!, Vol. 10, Oct. 1960.

29. Makarov, L. o., "Ultrasonic Welding" 1 A.kusticheskiy Zhurnal, Vol. 6, Oct.~c. 196o, pp. 507-508.

30. American Society for Metals, Metals Handbook, 8th Edition, Vol. 1.

31. Cannon-Muskegan Corp., "Rene 41", The Alloy SE!cialist, Series No. 86.

32. Allegheny Ludlum Steel Corporation, "AM-350 - AM-355", of Alleghenz Ludlum Precipitation.- Hardening Stainless Steels.

;:

33. Allegheny Ludlum Steel Corp., Special Steels.

34. Barker, J. F., et al, "Astroloy- a Superalloy for 19000f Usert, Metal Prost•••, I>ec. 1960.

35. E. I. duPont deNemours and Co. (Pigments Dept.), 11duPont Columbia. Alloy D-31", Technical Bulletin.

36. Fansteel Metallurgical Corporation, Tungsten.

37. General Electric Company, Tungsten at a Glance.

38. Grabecker, D. W. (ed.), Metals for Supersonic Aircraft and Missiles. A.m. Soc. for Metals, Cleveland, 1958. .. . .

39. Hampel,· Clifford A., (ed.) ,. Rare Metals Handbook. Reinhold, N. Y., 1954.

40. Jahnke, L. P., R. G. Frank, andT. K. Redden, "Columbium Alloys Today", June and July 1960.

41. Universal-Cyclops Steel Corporation, (Retractomet Division), "Mol7bdei!lla + O.$% Titanium Sheet•, Spec. No. MTS-59-9.

42. Aero projects Incorporated (Jones 1 J. B;yron, Nicholas Karopis 1 John G. Tho•s, and Dennison Bancroft), "Fundamentals or Ultrasonic Welding, Phase I" I RR..:59-105, Final Report on Navy Contract NOas-58-108-c, May 1959. Available f'rom OTS as PB 161677. · · ·

43. Aeroprojects Incorporated, (McKaig, H. L., and J. B. Jones) (CONFmENTIAL) 1 RR-60-94, Final Report on Navy Contract NOrd-18779(FBM), July 1960.

44. Aeroprojects Incorporated, Unpublished Work.

45. Aeroprojects Incorporated, "Applications or Ultrasonic Energtr I NY0-9588, First Quarterly Report on Task 5 or AEC Contract AT(30-l)-1836, Feb. 1§01.

56

46. Aeroprojects Incorporated (Jones, J. Byron, Nicholas Maropis, John G. Thomas, and Demison Bancroft), "Fundamentals of Ultrasonic Welding, Phase II", RR-60-891 Final Report on Navy Contract NOas-59-6070-c, Aug 1960.

47. Reuter, T., and R. H. Bolt, Sonics, J. Wiley & Sons, 1955.

48. Mason w. P., Piezoelectric Crystals and Their Application to Ultra­sonics, D. Van Nostrand Co., l9SO.

49. Am. Inst. of Physics, Physics Handbook, McGraw-Hill, N. Y., 1957.

50. Crawford, A. E., Ultrasonic Engineerir:g1 Academic Press Inc., London, 1955.

51 National Defense Research Council, Summary Tech. Report of Div. 6, Vol. 13, 1946.

52. Berlincourt, D., B. Jaffe, H. Jaffe, and H. H. A. Krueger, "Trans­ducer Properties of Lead Titanate Zirconate Ceramics", IRE Transactions, UE], No. 1, Feb 1960.

53. Erie Resistor Corp. (Oshry, H. I. and J. M.), "Physical Properties of Piezoelectric Ceramics"~ RP 30004, 1953.

54. Whymark, R. R., "Magnetostrictive Transducers with Mechanical Loads", Acoustica, Vol. 6, No. 3, 1956.

55. Dobelli, A. c., "Piezoelectric Transducers", Acoustica, Vol. 6, 1956.

56. Naval Ordnance Laboratory, (Nachman and Buehler), Tech. Report Navord

~· 57. Naval Ordnance Laboratory, (Davis and Ferebee), Tech. Report Navord 3947. .

58. International Nickel Co., "Design or Nickel Magnetostrictive Trans­ducer", Technical Publication, 1955.

59. Philadelphia Bronze and Brass Corp., Castings and Forgings, Philadelphia, Pa.

60. Ampco Metals, Inc., Ampco Metal Specification, Milwaukee, Wise.

61. Ampco Metals, Inc., Ampco Alloy Specification, Milwaukee 46, Wise.

62. Ampco Metals, Inc., "The Machining of Ampco Metal", Bulletin 66b, Milwaukee, Wise.

6). International Nickel Co., "Brazing and Soldering Nickel and High-Nickel Alloys", Tech. Bulletin T-34.

57

Wt4 x .u a; c UUA. 9R±

64. International Nickel Co., "Inconel X: Age-Hard enable Nickel Chromium Alloy", Tech .. Bulletin T-38 ..

65.. International Nickel Co., "Engineering Properties of K Monel and R Monel", Tech .. Bulletin T-9.

66.. Titanium Metals Corp. or .America, Titanium Engineering Bulletin No. 1., New York 7, N.Y.

67. Carnegie Illinois Steel Corp. (Camp, J. M. and c. B. Francis), The Kaki!!ih Shaping, and Treating of Steel, Fifth Ed., Pittsburgh, Pa.

68. Ramall, R. H., An Introduction to Acoustics, Addison Wesley Press, Inc., 1951.

69. Kolsky, H., Stress Waves in Solids, Oxford Press, 1953.

70. Climax Molybdenum Company (Freeman, Robert R.1 and J. z .. Briggs)., "Molybdenum for High Strength at High Temperatures".

71 Hunt, Frederick v., Electroacoustics, Wiley & Sons, 1954.

72. Neppiras, "Techniques and Equipment for Fatigue Testing at Very High Frequencies", Proc. or ASTM, Vol. 5.9 1_ .~959 ..

73. Von Heydekampr, Part 2-157 of Proceedl.ngs of the ASTM, 1931

74. Hanstock, R,. F., and A. Murray, "Damping Capacity and the Fatigue of Metals", J. Inst. Metals (London), Vol. 72, 1946, p. 97.

75. Thompspn, N., N. Wadsworth, and N. Lociat, ttThe Origin of· Fatigue Fracture in Copper", Philosophical Mag., Vol. 1, Serial 8, 1956, p. 113.

76. Mason, W. P., Physical Acoustics am the Properties of Solids, D. Van Nostrand Company, New York, 19SB.

77. Rayleig~, Lord, Theory or Sound, Dover Publications, Vol. 1, p. 281 · et seq. ·

78. Lamb, Horace, The DynamicalTheory of Sound, Edward Arnold, London, 1910.

1

I. 2

1

1

1

1

1

1

•

DIS'l'R!BUTiat LIST

Accoustica .Associates, Inc. Attnr R. J. Hurley, General Manager 10400 Aviation Boulevard Los Angeles 45, California

1

Aerojet-General Corp. 1 Attn: Kenneth F. Mundt, Vice Pres. Mfg. 6352 Irwindale Avenue Azusa, California

Aeronca Mf'g. Corp. Attn 1 L. C. Wolfe, Chiet Engineer 1712 Germantown Road Middletown, Ohio

1

AiResearch Manufacturing Co. 1 Attna Chief Engineer 4851 Sepulveda Blvd. Los Angeles 45, California

American Machine & Foundry Co. 1 Government Products Group Alexandria Division .Attn: J. D. Graves, General Manager 1025 North Royal St. .Alexandria, Virginia

Armour Research Foundation Illinois Institute of Technology Technology Center .Attnr Director, Metals

2

10 west 35th Street 2 Chicago 16, Illinois

Avco Corporation Nashville Division Attn: Mr. ,W. F. Knowe, Mgr. Design Eng. Nashville 1, Tennessee

Avco Corporation Nashville Division Attnl Mr. F. A. 'l'ruden, Mtg. Dev. Nashville 1, Tennessee

1

1

a

Avco Corporation ;R~search and Advanced Development Dlivision Attn• Director of Research , W11mington, Massachusetts

Beech Aircraft Corp. Attns Mr. E. Utter, Chief Structures Wichita 1, Kansas

Bell Aerosystems Company Attn: R. W. Varrial, Manager P:i-oduction Engineering P. 0. Box 1 Buffalo 5, New York

B. M. Harrison Electrosonics, Inc. Attnr Bertram M. Harrison, President 80 Winchester Street Newton Highlands 61, Massachusetts

Bendix Products Division Missiles Department Attnr Chief, Airframe Design Group 400 S. Reiger St • M1shawaka, Indiana

Boeing Company Attn: Boyd K. Bucey, Asst to Vice Pres-Mfg. P. 0. Box 3707 Seattle 24, Washington

Boeing Company Attn:: c. c. Lacy, Manager

Research & Development Aero-Space Division P. O. Box 3707 Seattle 24. Washington

Boeing Camp~y Attna Fred P. Laudan, Vice Pres.

, Manufacturing-Headquarters Office P. O. Box 3707 Seattle 24, Washington

Boeing Company Wichita Division Attnr w. w. Rutledge, Mfg. Mgr • Wichita, Kansas

1

1

1

1

1

1

1

I

1

• 1

Cessna Aircraft Canpany Attna R. L. Lair, Vice Pres & Gen Mgr

Prospect Plant Wichita, Kansas

1

Chesapeake Instrument Corporation 2 Attn a Director, Research & Development Shadyside, Maryland

Chrysler Missile Division Chrysler Corporation Attna Chief Design Engineer P. 0. Box 1919 Detroit 31, Michigan

Circo Ultrasonic Corporation Attn: Benson 'Carlin, Vice President 51 Terminal Avenue Clark, New Jersey

1

1

1 Convair Division of General Dynamics Carp Attn: R. K. May, Chief', M:f'g Res & Dev Engrg P. o. Box 5907 Fort Worth, Texas

Convair Division of' General Dynamics Corp 1 Attn: A. T. Seeman, Chief' of Mtg-Engr. P. o. Box 1011 Pamona, California

Convair (Astronautics) Division General Dynamics Corporation Attna J. H. Famme, Dir. of' Mfg. Dev. P. 0. Box 1128 (Zone 20-00 San Diego 12, California

Curtiss-'W'right Corp. Propeller Division Attn: J. H. Sheets, Works Manager Fairfield Road Caldwell, New Jersey

Curtiss-Wright Corp. Attn: H. Hanink, New Process Mtg. Woodridge, New Jersey

Douglas Aircraft Co., Inc. Attn: c. B. Perry, Plant Supv. 3855 Lakewood Boulevard Long Beach 8, California

1

2

1

b

Douglas Aircraft Co., Inc. Attna c. H. Shappell, Works Mgr. 3000 Ocean Park Blvd. Santa Monica, California

Douglas Aircraft Co., Inc. Attn: J. L. Jones, Vice Pres, Gen. Mgr. 2000 N. Memorial Drive Tulsa, Oklahoma

Fairchild Aircraft & Missile Div. Fairchild Engine & Airplane Corp. Attn: E. E. Morton, Mf'g. Technical Analysis Hagerstown, Maryland

General Electric Company Attn: Manufacturing Engineering Res Lab. Cincinnati 15, Ohio

General Motors Corp. Allison Division Attna N. F. BratkoviCh, P. o. Box 894 Indianapolis 6, Indiana

Gulton Industries, Inc. Attn: Walter Welkowitz

Supv. Joining Processes

Director, Research & Development 212 Durham Avenue Metchen, New Jersey

Harris ASW Division General Instrument Corp. Attnl Frank David, Chief' Engineer 33 Southwest Park Westwood, MassaChusetts

Lockheed Aircraft Corp California Division Attnt J. B. Wassail, Dir. of' Engineering Burbank, California

Lockheed Aircraft Corp. Missiles and Space Division Attn: Mr. Don McAndrews Supv. Manufacturing Research P. 0. Box 504 Sunnyvale, California

W' W%~

2

• 1

1

1

1

1

1

1

1

1

•

H:Donneil Aircraft Col'p. 1 Attn: E. G. Szabo, 'Mgr. Produ.ction Eng. Lambert-st. Louis Mo:nicipal Airport P. 0. Box $16 St. Louis .3, Missouri

Marquardt .Aircraft Co. Attn: J. M. Norris, Factory Mgr. 1 Box 670 Ogden, Utah

Marquardt Aircraft Co. ,Attna John S. Lie.teld., Dir. of Mtg. 16555 Saticoy Street · Van Nuys, Calif •

The M~rtin Canpany

1

•:ttn: Chief Engineer 1 P. o. Bax 179 Baltimore .3, 'Maryland

The Martin Company Attnt .. Chief Librarian, Eng. Lib. Baltimore .3, Maryland 1

The Martin Company Attnr. L. J. Lippy, Dir. Fabrication Div Denver, Colorado

North American Aviation, Inc. .a;ttn1 Chief Eagi:D.eer Port Columbus Airport Columbus 16, Ohio

1

North American Aviation, Inc. 1 Attnt Latham Pollock, Gen. Supv. Mfg. Bng International Airport Los .Angeles 45, Calif.

Northrop Aircraft, Inc. Attnt R. R. lol&Jl, Vice Pres. Jff'g. 1001 E. Broadwa;r Hawthorne, Calif'.

Northrop AircrJtt, Inc. Norair Di,vision Attn: Ludwig Roth, Dir., Research

Engineeriilg Department 1001 E. Broadway . Hawthorne, California

c

1

1

Pratt &. Whitney Aircra.t"t Div. United Aircraft Corporation Attnr L •. M. Raring Chief, Metallurgical &. Chemical Lab :P.o. Box 611 Middletown, Conn.

Republic Aviation Corp. .Attn: Adolph Kastekowits, Chie.t" Kf'g. Engr. Farmingdale, Long Island, lew York

Rheem. Mf'g. Company Aircraft Division Attn 1 Chief' Engineer 11711 · S. Woodruff Ave. Downey, Calif'.

Rocketdyne Division North American Aviation, Inc. Attn: R. J. Thompson, Jr., Dir. Research 66.3.3 Canoga Avenue Canoga Park, Calif.

Rocketdyne Division North American Aviation, Inc. Attnr Mr. J. P. McNamara, Plant Mgr. P. O. Box 511 Neosho, Missouri

ROhr Aircraft Corporation Attnr. Chief' Structures Engr. P. O. Box 878 Chula Vista, Calif'. .

Rohr Aircraft Corp. ~ttn: Burt F. Raynes, Vice Pres. Mtc. P. o. Box 878 Chula Vista, . Calif'.

.Ryan Aeronautical Com.pan;y . Attnt Robert L. Clark, Htg. Works Mgr. Lindbergh Field San Diego, California

Sciaky Bros., Inc. 4915 W. 57th Street Chicago .38' nlinois

10 Armed Serviees Technical ID.toiW.ticm .&,enq Attnt DoctlliSnt Service Center (!IOSCPJ Arlington Hall Station .Arlington 12, Virginia

6 & Aeronautical Systems Divisiaa 1 1. repro. Attns Maauf'acturing Teoh:aology I.ab(ASBCJT)

Wright-Patterson Air Force Base, Chio

•

1

1

I

Air Force Systems CODili&Dd. Attnt Kr. C. W. Inii"fin (R.llR.AB-F) Andrews Air Force Base, M.ar7land.

Aeronautical Systeas Division .Attnt ASRimB ~ight-Patterson Air Force Base, Ohio

1

1

2 Aeronautical Sy~tems Division . . . Attn: Metals & Ceramics Lab (ABRCM) wright-Patterson Air Force Base, Ohio

1

2

1

1

1

1

Aeronautical Systems Division Attn: Appiications Lab {ASRCE, Mr. Teres) WTight-Patterson Air Force Base, Ohio

Aeronautical Systems Division

1

Attn s fiight Dynamics tab. 1 Structures Branch (ASaMDS)

wright-Patterson Air Force Base, Ohio

Battelle Memorial Institute Defense Metals Information 1 Attnr Mr. c. s. Dumont ~0~ King Ave. Columbus, Olio

Ballistic Missile Syat.-a Division Attn: Industrial ltesourcea P. O. Box 262 1 AF Unit Post otfice Inglewood, Calif.

Chief, Bureau of Naval weapons (P!D-2) Department of the Bavy Washington 2.5, D. C. 2

Frankford Arsenal Research Institute, 1010 {11:0-1) !ttnr Mr. E. R. Rechel, Deputy.Director Philadelphia 37, Pa •

d

l Solar Aircraft Company Attnt Engineering Library 2200 Pacific Highway · San Diego, California

Temco Aircraft Corp. Attn: D. T. Brooks, Mfg. Mgr. P. o. Box 6191 Dallas, Texas

Sout:q.west Research Institute .Attn:r Glenn Damewood, Dir. Applied Ph7sics 8 ~00 Cule bra Road Dept. San Antonio 6, Texas

Union Ultra-sonics Corporation Attn: John Zotos, Chief Project Scientist lll Penn Street Quincy 69, Massachusetts

Vought Aeronautics Division Chance-Vought Aircraft, Inc. Attns George Gasper, Mfg. Engr. Mgr. P. 0. Box ~909 Dallas, Texas

Vought Aeronautics Division (fuance-Vougb.t Aircraft, Inc. Attnr Chief Librarian, Engineering Library Dallas, Texas

Vought Aeronautics Division Chance-Vought Aircraft, :me. Attn a J. A. Millsap, Chief Engr.

Manufacturing Research Development p •. o. Box .5907 Dalia a, Texas

G. 0. Marshall. ·Space Flight Center National Aeronautics & Space Administratian Attn 1 William A. Wllson,

Chief, MR & D Branch Huntsville, Alabama

Langle7 Research Center National Aeronautics & Space Administrative Attn: Technical Director Langley, Virginia

2 Commanding General Redstone Arsenal Rocket & Guided Missile Agency Attn: Chief, Space Flight Structure Design Redstone Arsenal, Alabama

e

AD Aeroprojects Incorporated, West Chester, Pa.

DEVELOPMENT OF ULTRASONIC WEIDING -EQUIPMENT FCR REFRACTCRY ME'tALS, by J. B. Jones et al. September 1961. 66p. incl. illus. tables. (ASD Project 7-888) (ASD TR-7-868{r)) {Contract AF33(600)-43026) .

Unclassified report

Joining refractory and superallo;y metals in thicknesses up to 0.10 inch by u1 trasonic welding is feasi­ble. A first approximation of the

(OV'er)

UNCLASSIFIED

l.Materials--Proeessing

2.W81d.1Dg 3.Met8ls I.Jones, J. B.­II ~Aeroprojects-

IncoJ"Porated m.contract ... AF33{600)-43026 IV.ASD Project

7~888 V.Fabrication

.Branch

AD ... Aeropro jects rncorporated, West 1 ~t~rials:"" Chester, Pa. ~Probessing

DEVELOPMENT OF ULTRASONIC WELOING - 2. lreldiilg EQUIPMENT Fal REFRACTCilY METI,Ls, b,1 3.Metala J .. B. Jones et al. September 1961. I~Jc:>nes~ J~ :9.-.-66p. incl. illus. tables. · (ASD II.Aerop~jects-Project 7-888) . (ASD TR~7-888{I)) . ·xncorparated. {Contract .AF33(600)-43026) _ · · III~Cont:ract _ ·

Unclaasf..!~d report . AF3~(6oo)-43026

AD Aeroprojects Incorporated, WBst .. Ohester, Pa.

DEVELOPMENT OF ULTRASONIC WELDING EQU:ImENT F<R REFRAC'tCRY .JmlrALs, by . J. B. Jones et al. SBptember 1961. 66p .• incl. Ulus. tablese (ASD. Project-7-888) {ASD TR-7-888(!)) (Contract AF33(6oo)-43026) ._ ..

Unclassified report

Joining refractory and superall07 metals in thicknesses up to 0.10 ineh by Ul trasoid .. o ·welding -is feasi­ble. A first approximation ot the

(over)

.AD ~~roprojects Incorporated, West Qhester.~ Pa •. I)EVELO~ OF ULTRA:SONm. wELDOO -EQUIPMENT FOR J~EFRAOTORY. ~At,$_, __ by J. a. Jones et ~l. Septem~r_I~6I. ~6p. incl. illus. tables. ,(ASD Project ~7-8_ 88) (_.\sD TR-7-888{l)) (Contract AF33(6~430~6) ·~·~

. UI\.Classitied reiJort IV ~.ASD Project·

Joining refractory and nperall:oy · . 1 ~888 Jo~ning retracton and super alloy . metals in thickneaaes up to O.IC-- V .l"abricaticm DJBtals in thickneaaes up to 0.10 inch by ultrasonic wld:hlg is !eaai- Branch inch by ultrasOnic welding is :f'ead.-

1' ~ • A first approximation ot "the ble. A first appraxiJDB.tion ot ·1ohe

UNCLASSIFIED

!.Materials-Processing

2.Welding 3.Metals !.Jones, J. B. II.Aeroprojects . . Incorporated IIIoContract

AF33(600)-43026 IV.ASD Project

·7~88 VoFabricatian

Branch

UNCLASSIFIED

!.Materials.;. ·Processing

_2.~ldlilg 3.Metala I.Jones, J. B~­.ri,Aeropr()jej:ts ~ . . -~corpora ted rrr.oontract ... AF33(6oo)~43o26 IV •• SD Project

-7~8 V.~Fabrica~i.on

~Branch

tJNCU:SSD'DD ; (over) URCL.lSSirlED Cover) ~---------------------------L----------~--L-------------~--~------~----~--------~