Ultrasonic welding

67 %
33 %
Information about Ultrasonic welding

Published on February 14, 2014

Author: aminezazi5

Source: slideshare.net


Ultrasonic welding

ULTRASONIC WELDING by H. P. C. DANIELS* After a general introduction and a description of the apparatus a survey is given of the variables involved in ultrasonic welding, in particular the pressure, the power and the welding time, and also their mutual dependence. It is shown that the physical properties, and the dimensions and the state of the surface of the materials under test also play an important role. Most measurements have been carried out on aluminium and copper. There is not enough data available to give a detailed explanation of the mechanism of ultrasonic welding. When the welding conditions are known, however, a reproducible weld can be obtained. These investigations have been confined metal welding to l J u r i n g the last ten years ultrasonic welding has become increasingly versatile. Although the early work was mainly confined to soft aluminium foil, it was soon realised that the process could be used successfully in joining such materials as copper, nickel, titanium, zirconium, and some of their alloys. Attempts to weld combinations of dissimilar metals also resulted in bonds of surprisingly good quality and strength. It is also possible to join metals to non-metals, e.g. met',d leads to germanium, silicon and glass. APPARATUS resonance frequency of the whole vibrating system. Since most generators require a real, or at least an approximately real, load they should be tuned more or less continuously to the varying resonance frequency of the acoustical system under operating conditions. In our experiments a feedback system was used in two of the spot-welders to automatically adjust the generator to the momentary resonance frequency el the vibrating system. A design of an ultrasonic seam-welder is given in Fig. 2. In this welder the whole vibrating system rotates, i.e. the piezomagnetic transducer as well as the amplitude transformer. The anvil also rotates, but in the opposite direction, and at the same peripheral speed as the welding tip. The vibration is therefore perpendicular to the direction in which the workpieces are moving. Four different set-ups were used, as shown in Table 2. There are several designs of ultrasonic welder. One ol these, namely a spot-welder design for welding metals, is schematically given in Fig. 1. The electric power of an electronic generator is fed to a piezomagnetic (magnetostrictive) transducer, which converts the a.c. electrical power into a mechanical vibration. Since the excursion amplitude of the longitudinally vibrating transducer is too small, a mechanical amplitude transformer is used to increase its amplitude. The extreme end of this transformer acts as a welding tip. There are several shapes for this transformer, t e.g. bicylindrical, exponential, conical," catenoidal ~ and GaussianJ, 3 They all give different amplitude transformation ratios and have different stress distributions. It must be noted that the Gaussian transformer, in contrast to the others (especially the bicylindrical), shows a constant stress amplitude along the axis. The transformation ratio is not the only consideration in the design of a transformer if high intensities are required. Obviously the m a x i m u m velocity amplitude at a welding tip is limited by the dynamic strength of the transformer material and consequently by the stress distribution inside the transformer. Typical transformation ratios are given in Table I. In the present experiments a bicylindrical transformer was used. As in all metal welders a static clamping force is exerted in a direction normal to the transformer axis, i.e. normal to the direction of vibration. Since the welding tip can slip along the first material and/or since the second material can slip along the anvil, energy can be dissipated at interfaces other than the one between the metals being welded. But by choosing suitable metals for the anvil and the welding tip this energy loss can be limited, We used a welding tip of titanium alloy I.C.I. 318 A and an anvil of hardened steel. Loading the system changes the imaginary part of the acoustical impedance and, consequently, the mechanical Velocity in axial direction * Philips Research L a b o r a t o r i e s Eindhoven-Netherlands Fig. I. Ultrasonic spot welding a p p a r a t u s 190 N.V. Philips' VARIABLES IN ULTRASONIC WELDING The action of ultrasonic welding is as follows. Two workpieces, which are clamped between the welding tip and the anvil, are rubbed against or along each other by the vibrating tip. This vibration causes friction at the interface between the two workpieces. The friction cleans the contact surface by pulverizing and partly removing contaminants and oxide layers, and contributes to the formation of small spot-welds which grow into a weld. Apart from frequency and ambient temperature which are usually more or less fixed, there are three important variables which can influence the quality of the weld, Piezomagnetic transducer # ) II t~ DI II II ~ I ~ i÷********b ~ d [ 14 L ~ / ~18.5// Welding tip Gloeilampenfabrieken ULTRASONIcs/October-December 1965 [ o Power amplifier I I (generator) ~ -~2o~c/s ,,|Piezoelectric pick-up Veloc,ty transformer (iyidio) bclnrcl Workp~eces

Table 1 TRANSFORMATION RATIO FOR VARIOUS TRANSFORMER SHAPES Transformer shape 1200 ,1 Velocity transformer ratio Bicylindrical N2 Conical N(cos kl--NklSinkl ) N-I Exponential N N. Catenoidal ( d-2-dl = cash 71 j Gaussian 4- A/4 cylinder = input diameter d., = output diameter = length / 1 -- cos k,l (1 + 2 In N2)1''-' / dl I000 N= .~ r n BOO C ¢ Copper-nickel (45-55) Thickness0-3ram Weldingtime O.Ssec dl ~{~ , k = 2rr ' k/ = 2rr X ~' ,2 k s = (k") 2 4- 72 u 600 ¢o £ Table 2 ~ J ULTRASONIC WELDERS USED IN THE AUTHOR'S EXPERIMENTS Copp?~ +2 Maximum electrical power (W) Type of weld i Frequency [kc/s] Seam weld 20 100 40 2000 Spot weld 35 600 Manufacturer ._E 4 0 0 Motional feedback Philips, Netherlands Sonoweld, U.S.A 20 Mullard, U.K. Yes 20 Lehfeldt, Germany .A Yes No No ~ 200 ~ . 0 namely the clamping force, the ultrasonic power, and the welding time. These variables must be adjusted to the right values for the various combinations of materials, dimensions and shapes of the parts being welded. The clamping force F is necessary for an intimate contact between the workpieces, so that the vibration energy can be delivered to the surface to be welded. When the ultrasonic power P is changed during welding, the vibratory amplitude is changed, and hence also the dynamic stress in the weld surface. The power should not be too high nor too low. I f power is too high, the large AI mini m 50 I00 Clampingforce, F [kg] 150 Fig. 3. Relation between the lower limit of the electrical power and the clamping force for aluminium, copper and copper-nickel dynamic interracial stresses will damage the weld; it it is too low,, no weld will be made. It is generally found that thick or hard materials require more power than thin and soft materials. Since energy is the product of power and the third main variable, welding time, and since the thickness and hardness primarily determine the amount of Velocity t in axial direction Clampingforce Exponentialvelocity i~___~ transfo__rmer We tip W 1Piezomagnetic " transducer = --- ×/ Anvil [l Fig. 2. Ultrasonic welding machine . . . . continuous seam- . -== Generator --~20kc/s N Bail bearings c ULTRASONics~October-December 1965 191

energy that is required, the welding time has also to be taken into account. The frequency can be chosen over a wide range, from below 100 c/s to more than 100,000 c/s. For practical reasons it is chosen between 15 kc/s and 45 kc/s. In the following discussion a weld will be defined as good if it does not fail when a nugget is torn out in the peeling test. Such a good weld can only be realized when the above mentioned variables have the right values. What those values are depends on the physical properties of the materials under test and on their surface state. The variables are dependent on each other. A few of these inter-relations w i l l now be discussed. The lower limit of the electrical power P necessary 800 f Copper (annealed) ] Welding time I 0 sec P= f(F) (limiting values) - - --Fop t= 6 2 + 1 8 0 t I o._ 6 0 0 ,2 o G B E 4O0 u ~J ;200 - make a g o o d weld depends on the clamping force F, when the other variables are kept constant. This relation is shown in Fig. 3 for different metals with the same thickness and in Fig. 4 for various thicknesses of copper. The solid lines in Fig. 4 give the limiting power below which no good weld could be obtained. From these graphs it is seen that each F-P curve shows a minimum. Moreover the dasheddotted line passing through these minima in Fig. 4 points to a linear relation between thickness and the optimum clamping force F,,~,. The relation for copper-to-copper welds with a sheet thickness t of up to 0.3 mm is found to be F ..... 62 180 t (kg) Moreover it is interesting to note that Fig. 4 enables the relationship to be established between the minimum power for a good weld P , , , and the thickness t. It is found to be a proportionality as shown in Fig. 5. The minimum electrical power necessary for a weld depends on the construction of the apparatus, the efficiency of the transducer and also on the matching between the transducer, the transformer and the load on the one side and the electrical generator on the other side. It should be noted that in the welding of materials with different thicknesses, the thinner material must be placed nearer the welding tip, in order to limit the energy dissipation in the thicker member and to prevent strong deformation of the thinner plate, which is usually observed when it is placed nearer the anvil. In Fig. 6 the breaking force is given for aluminium welds as a function of the welding time for a given electrical power and clamping force. From this one concludes that: 1. The breaking strength is little lower than the tensile strength. The breaking strength for the aluminium weld having a weld area of approximately 9.5 mm" is found to be about 8.5 kg/mm 2, whilst the tensile strength of the aluminium used in the present experiments is 10 kg/mm." 2. The breaking force does not decrease at longer welding times (greater than 5 sec). Consequently, metal fatigue must be ruled out for aluminium. 3. There is a peak in the curve at a welding time of about 0.8 sec. This was also found by Weare a but up to now there has been no explanation of it. 4. The build-up of a weld is a continuous process. ~ g ULTRASONiCs~October-December 1965 ,/! i _ Fopt IO0 Fopt 150 Clamping force, F [kg] 50 J 2~0 Fig. 4. R e l a t i o n b e t w e e n the l o w e r l i m i t o f the e l e c t r i c a l po~cr a n d the c l a m p i n g force for c o p p e r o f d i f f e r e n t t h i c k n e s s e s , n a m c l ) 0.06, 0"1, 0.2 a n d 0"3 m m 4o0[ opper II~ Welding annealed time I-Osec i I Clamping force Fopt = 62 + 180t T o E 200-- E 2 J i Z [ ! J ± 0~ 0 0 I 02 0.3 Thickness, t [mm] Fig. 5. ( A b o v e ) M i n i m u m electrical p o w e r r e l a t e d lo t h i c k n e s s o f c o p p m r i 6O ! Aluminium (TI36) i Thickness 0 ' S m m Clamping force 8 0 k 9 7 Electrical power 2 8 0 W °~o i I CONDITIONS FOR OBTAINING A GOOD WELD in the preceding section it was shown that obtaining a good weld depends on the power, the clamping Force, and the welding time, but these in turn depend on the physical dimensions and properties of the material to be welded. This will now be discussed in some detail. f.=O'06 rr~m %~4 0(~ 2o~- 192 - to -- i i [ ! t 1 I s Fig. 6. _ ~o I i ~5 20 Welding time [sec] B r e a k i n g force as a f u n c t i o n o f w e l d i n g t i m e i 1 1 25 J 3o

Physical dimensions The relations between the thickness of the metal sheet and the minimum power as well as the optimum clamping force have already been illustrated in Fig. 4 and 5. In addition Fig. 7 shows the influence of the length of the metal strips. At the far end the strips were blocked in a pair of tongs and at the other end they were clamped as usual between the welding tip and the anvil. The distance between the blocked end and the welding tip was varied from 2 m m to 48 mm. It is found that the quality of the weld oscillates as a function of the strip length between the blocked and the clamped ends. This is probably connected with oscillations in the energy dissipation in the strips because of mechanical resonances at some lengths of strip. IO0 o 80 ~" 0"3 uponl6owO.Dmrn [ ~ 60 Material properties '~ 4O The influence of the kind of material has already been demonstrated in Fig. 3. The hardness of the metal is also a very important factor. The harder the metal the higher the minimum power to make a weld. A weld between metals with a different hardness needs a minimum power which is about the minimum power necessary for a weld of the harder metal (Fig. 8). Moreover, the adjustment of the clamping force is often more critical. ~ 20 Clamping force 20kg | Welding t i m e Isec Width 10"7 m m ThickneSs 0.3 and 0"5 m m __ Surface condition The condition of the surface of the workpieces and of the anvil is of great importance. Curves b and c of Fig. 9 show the breaking force for aluminium welds, the contact surface of which had been polished or lightly roughened, respectively, before the welding experiment. The breaking force with ground surfaces is smaller than with polished surfaces. In the former there has been a smaller relative displacement between the workpieces, so that less frictional energy has been delivered to the contact surface. Consequently the best results are obtained with workpieces having polished contact surfaces and rough surfaces on the opposite (tip and anvil) sides. The scattering of the experimental values for the case where the workpiece surfaces had been ground (curve c) should probably be ascribed to the bad reproducibility of the surface roughness. Roughening the anvil by sandblasting improves the results, as demonstrated by curve a of Fig. 9. 20 30 40 Distance b e t w e e n the blocked end and the welding tip [ m m ] 50 I0 Fig. 7. Relation between the weld size and the length of the workpieces Fig. 8. Welding of similar and dissimilar metal sheets. First m e n t i o n e d sheet of each pair is placed at the tip side, second sheet at the anvil side 8OO n -| #pper 1 1Aium[nium o~600 Q. "VCopper aluminium 1_ ~400 [ I Sheet thickness 0.3 turn Welding time I sec k Copper-copper 4~ Cleaning Cleaning the workpieces is not very important and may be omitted provided the contamination is not too severe. For instance, lacquer-coated copper wire can be welded to a copper plate without the coating of the wire having been previously removed. However, clean surfaces give a better reproducibility, so that prior cleaning is recommended? ,'5 MECHANISM OF ULTRASONIC W E L D I N G At the start of the welding process there is a reciprocating sliding friction of the contact surfaces of the workpieces. During this initial stage the surface layers of contaminants and oxides are pulverized and partly removed. Owing to the friction in the contact area, the temperature will rise. In our experiments the temperature of the contact area was determined by measuring the thermoelectric e.m.f, between the workpieces during welding. The experimental maximum temperatures were not higher than about 40 % of the melting +z_ ~ 2 O0 - - ~ . , , ~ ~ o~Pmin ._1 ] o I ma[urninium I 0 Fopt 50 1O0 Clamping force, F [kg] 150 200 point of the metal expressed in degrees centigrade. Similar values have been found by Weare, 4 Jones, ~ Baladin 6 and Okada. 7 It should be noted that it is extremely difficult to obtain meaningful data, since the thermoelectric e.m.f. represents the average temperature for the whole contact area and not the maximum temperature occurring at small spots during welding. In friction experiments, Bowden and Tabor s have found high temperature peaks at small spots. It is interesting to note that in the present experiments cross-sections of ultrasonic welds did not show melting ULTRASONics~October-December 1965 193

phenomena. Moreover, under a micro-scan analyser, with a resolution of about I am, no diffusion could be observed on a copper-to-nickel weld. Okada 7 who also used a micro-scan analyser, was not able to detect diffusion in the interface either. Therefore the mechanism is unlikely to be explained as a melting process. We believe that the explanation should be sought in a process of plastic deformation. We shall describe this idea briefly. Owing to the sliding friction the contact surface is cleaned. In this cleaned surface small spot welds are made. The number of small spot welds on the cleaned surface increases by the joint action of the steady clamping force and the alternating tangential vibration. At some point during a weld cycle this friction process mnst stop, otherwise the initial spot welds will be destroyed by the relative m o v e m e n t of the workpieces. A steady state is reached in forming and destroying small spot welds. This alone would imply that the breaking force of a weld is much lower than the tensile strength, which is contrary to what is seen in Fig. 6. Therefore another mechanism, quite distinct from sliding friction, must complete the weld. The growth of the initial junctions must be ascribed to the combined normal and shear stresses (see also Bowden and T a b o r P Since the deformation decreases with increasing hardness, under the same shear stresses, it is easy to see that harder metals need more energy. APPLICATIONS Welding of metal to metal Welding of sheets. Ultrasonic welding apparatus makes possible the welding of similar and dissimilar metal sheets. 80 l In Fig. 10 a review is given of weldable and unweldable materials and combinations of materials. Both metal to metal and metal to non-metal joints can be made. The latter are discussed in the next section. In the welding of dissimilar metal sheets, e.g. copper to aluminium, it makes a difference whether the aluminium or the copper sheet is placed nearer the welding tip. This is shown in Fig. 8. Moreover, the same graph clearly indicates that the copper-to-aluminium and aluminium-to-copper welds require minimum electrical power, whereas the optimum clamping force F ..... is lower than the F,.,,, for aluminium-toaluminium and copper-to-copper welds. In addition it is demonstrated that the welding of dissimilar metals is more critical. Sometimes, the position of the workpieces is important to prevent wear between the welding tip and the sheet metal adjacent to it. For instance, in the welding of titanium to aluminium with a titanium welding tip, it is necessary to place the aluminium next to the tip. Welding o[" wires. It is possible, with various metals, to weld wires to plates or sheets as well as to other wires. To weld a wire to a plate, a groove can be made in the welding tip in order to improve its grip on the wire. This is sketched in Fig. 1 1. It was found that the initial strength of the wire was almost retained when both ends of the groove were curved upwards. See Fig, liD. WeMing of wirex upon evaporated metal layers. It is possible to weld wires upon metal layers evaporated on a glass substrate, provided the adhesion of the layer to the glass is satisfactory. No disturbance of the layer was observed when aluminium, copper or gold wires were welded to nickel-chromium-nickel layers evaporated on a glass substrate. The thickness of the layer was about I Sandblasted i ~ polished aluminium l GP;llsUhe%ConlVLinium E E u~ 60 E E~ E 2 ._~m o0,J Z 4o C )m@'-_ x : 2:1 a ~ ~o=o__ --@LOO ~@ <~ m m u U t.c L9 C9© ~ Ground anvil smooth ground atumlnium o 6 E l~lllololololo o o O C C-- u ~ . _ q) -->+OOq# O . . . . O D o/o o o l l l l o i - - F l ~ l l [ l l ~ t o l o l I I I I l[llllt lli o ~ o ~llHol I IIII]1 IIIIl~lllll ~ll],l, %°o o111t b I i o,, o • Aluminium TI36) 0"8 mm Clamping force 80kg Welding time O-Ssec ,toi id Iol o ,,!o! !,!i IIIIU [ I~tllll I ~t I °o Io11III1I,,,lill I 1ol II I IIIIII IIII Copper Copper-nickel Fernico Germanium 0 200 400 E l e c t r i c a l power [W] 600 800 Fig. 9. The influence o f the surface state o f workpieces and anvil on the .b~caking force o f a weld ULTRASONics/October-December 1965 Glass Gold Lead Magnesium Nil M o l y b d e n u m I I Nickel I Niobium Palladium I IPlatinum SiFcon I I (I Illlll joj III S i l v e r 1o liIIIIlIIII I IStainl, s t e e l o IIII i;¢,1 I I~111111 IIII S t e e l Iol ] I01 194 111111Aluminium ] IBeryllium III I IBrass M S 5 8 I • m 0 O . . . . ~ Welded in present e x p e r i m e n t s Welded in G e r m a n y Welded in U.S.A. Not i n v e s t i g a t e d Unweldable titl n o w Tantalum Tin Titanium Tungsten Vanadium Zinc Zirconium Fig. IO. Joints m a d e with ultrasonic welding a p p a r a t u s up to Spring 1965

~ . . . Fig. 11. The shape of a grooved welding tip for welding wires on plates . c A D Wire before w e l d i n g - - ~ Wire o f t e r welding Cross-section o-b 0.5 `am. Inspection t h r o u g h the b a c k o f the glass did not reveal a n y d a m a g e to the layer. Table 3 gives the shear strength for these metals welded to n i c k e l - c h r o m i u m - n i c k e l layers. Table 3 W I R E THICKNESS, MINIMUM ELECTRICAL POWER, AND SHEAR STRENGTH FOR SOME METALS WELDED UPON A (NI-CR)-NI LAYER EVAPORATED ON GLASS. W E L D I N G TIME 0 " 4 5 SEC. Metal Wire thickness (tzm) Minimum electrical power (W) 150 100 100 1"5 1"5 2 A1 Au Cu vibration Shear strength (kg/mm 2) > 2* 1"5 2 1"7-2 * On loading the aluminium wires after welding to the evaporated metal layer, the wires broke at the deformed edge of the weld at a load on the weld of 2 kg/mmL Welding o f m e t a l to n o n - m e t a l As mentioned already, not only metal to metal welds but also metal to n o n - m e t a l welds can be made. Table 4 gives a survey o f some welded materials and the thicknesses in question. It m u s t be n o t e d that this is neither the m i n i m u m n o r the m a x i m u m thickness. M o r e o v e r these metals have been chosen quite arbitrarily. The welds were m a d e with the 100 W and the 35 W spot welders m e n t i o n e d in Table 2. Since the spot welders have not the same electroacoustic efficiency, the relation between the p o w e r per unit area of the weld and the thickness is not linear for the a l u m i n i u m wire silicon welds. The 200 `am a l u m i n i u m wire was welded with the 35 W spot welder, which has a high-efficiency ceramic ferroxcube transducer, quite opposite to the low efficiency l a m i n a t e d metallic transducer o f the 100 W spot welder. The d a t a of Table 4 clearly d e m o n s t r a t e s that it is m o r e difficult to weld an a l u m i n i u m wire t h a n an a l u m i n i u m foil to n o r m a l plate glass. In a d d i t i o n a l u m i n i u m wires and gold wires were welded on e v a p o r a t e d layers o f gold, a l u m i n i u m and nickel having glass as a substrate. There was a g o o d electrical contact between the two metals, a l t h o u g h inspection t h r o u g h the back o f the glass revealed that the welds on e v a p o r a t e d gold a n d a l u m i n i u m layers were m o r e like welds between the metal wire and the glass. This did n o t pertain to the e v a p o r a t e d nickel layer. C o n sequently, it is possible to weld metal wires to an e v a p o r a t e d a l u m i n i u m or gold layer, but owing to the small a d h e s i o n between the e v a p o r a t e d metal layer a n d the glass, the metal layer u n d e r n e a t h the weld is removed when the necessary shearing force is a p p l i e d d u r i n g the weld cycle, so that the metal is welded to the glass. The c o n t a c t wires of silicon a n d g e r m a n i u m transistors can be welded to electrodes e v a p o r a t e d on to the semiconducting material. Experiments were also carried out on transistors without e v a p o r a t e d electrodes. F o r instance, a l u m i n i u m and gold wires were welded successfully u p o n the emitter and base of a p - n - p silicon transistor. Decreasing the thickness of the p - t y p e layer f r o m 5 a m to 0"5 `am on t o p of a thick n-type silicon matrix does not disturb the transistor properties. C o n s e q u e n t l y ultrasonic welding does not introduce severe strains or dislocations. W h e t h e r there is an e v a p o r a t e d metal layer between the wire and the silicon matrix or not, the p-n j u n c t i o n is n o t disturbed. In the present experiments the thickness of the wires was: aluminium, 80 `am and 100 a m ; gold, 15, 25, 70 a n d 100 `am. Examples of miscellaneous a p p l i c a t i o n s including both metal to metal to non-metal welding are illustrated in Fig. 12. It should be noted that b o t h similar (Fig. 12n) a n d dissimilar materials can be seam welded. However, the electroacoustic efficiency o f seam-welders is smaller than Table 4 K I N D OF MATERIAL, THICKNESS CHOSEN, MINIMUM ELECTRICAL POWER, AND OPTIMUM CLAMPING FORCE FOR SOME METAL TO NON-METAL WELDS MADE WITH THE | 0 0 W (a) AND 35 W (b) SPOT WELDERS DESCRIBED IN TABLE 2 Material Thickness (micron) i Tip side A1 1 AI AI Wire u AI | AI ) AI foil [ Anvil side Ge Si siSi Fused quartz Soda glass Soda glass Tip side 200 200 80 100 200 200 100 Spot welder Anvil side i 1000 4000 4000 4000 2000 2000 2000 b b a a b b b Minimum electrical power per unit area ( W/mm 2) Optimum clamping force (kg) 100 140 570 470 325 4500 100 300 165 215 35 -40 4OO 4OO 2900 ULTRASONics/October-December 1965 195

lcm 0.5cm lcm t Fig. 12. Examples o f cellaneous applications A Tin-plated copper lcm ! I miswire welded upon aluminium foil o f 10tzm by four spot welds B Plastic-coated copper wire welded upon copper printed ,/ circuit C, D Plastic-coated copper wires welded upon copper wires and tin-plated copper wires E Gold wires welded upon evaporated nickel layers (0-5 Fro1 on a glass substrate F Gold wires (18/~m diameter) welded upon a l u m i n i u m layers evaporated on a silicon substrate. Also note welds upon leads o f the transistor D G Thermocouple welded upon a glass substrate. The three parts have been joined simultaneously. So far only a l u m i n i u m and nickel wires can be welded upon glass H A l u m i n i u m (60 /,rn thick) welded upon aluminium 150 tLm thick. Seam weld E G F 1 I I lcm that of spot-welders owing to the geometrical limitations imposed by the rotating velocity transformer and welding tip system. The welding-tip has now the form of a flexural disc resonator which is loaded at one point only, as shown in Fig. 2. In this case the power transfer from the resonating welding-tip to the workpieces depends more critically on the kind of acoustic loading, viz. the kind and the thickness of the sheets and the magnitude of the clamping force. CONCLUSIONS As to the practical results of the ultrasonic welding experiments described in this paper, the following conclusions can be drawn : 1. Similar and dissimilar metals can be welded. 2. The minimum electrical power required at the optimum clamping force is proportional to the thickness of the sheets, and depends on the kind of metal being welded. 3. The optimum clamping force is linearly related to the thickness t of the sheet, viz. F .... ~ i /~t. 4. Owing to various mechanical resonance frequencies of the workpieces it is often necessary to block the workpieces, in particular sheets, at some optimum distance from the useful clamping point. 5. The surface state, in particular the degree of roughness, is of great importance. Polished contact surfaces and roughened surfaces on the tip and the anvil sides of the workpieces are recommended. 6. Cleaning the contact surfaces is not necessary though sometimes recommended. 7. Welding wires to sheets or wires requires the use of a groove in the welding tip. 196 / October- December 1965 ULTRASONICS, I I 0"2cm I 1 cm t ! lcm 8. Wires or foils can be welded to metal layers evaporated on glass substratcs, provided the bond strength between the layer and the glass is sufficiently strong. 9. Welds can be made between metals and non-metals. e.g. metals to silicon or germanium. ACKNOWLEDGMENTS The a u t h o r is indebted to Dr. Th. P. J. Botden and C. M. van der Burgt for discussions and helpful comments on the manuscript. Thanks are also due to M. Klerk for carrying out micro-scan analyses. REFERENCES I. NEPPIRAS, E. A., "'Very h i g h e n e r g y u l t r a s o n i c s , " British Jour,al o[" Applied Physics, II, 143 (April 1960). 2. MErKULOV, L. G., " D e s i g n o f u l t r a s o n i c concentrators," Soviet Physics-Acoustics, 3, No. 5, 246 (1957). 3. KLEESATTEL, C. " V i b r a t o r a m p u l l a c e u s , " Acustiea, 12, No. 5, 323 (1962). 4. WEArE, N. E., et al., " ' F u n d a m e n t a l s t u d i e s o f u l t r a s o n i c welding," Welding Journal, 39, No. 8, 331 ( A u g u s t 1960). 5. JONES, J. B., et al., " ' P h e n o m e n o l o g i c a l c o n s i d e r a t i o n s in u h r a sonic welding," ibid., 40, No. 7, 7 (July 1961). 6. BALADIN, G. F. a n d SILIN, L. L., " M e t h o d s for o b t a i n i n g steady c o n d i t i o n s in the u l t r a s o n i c w e l d i n g o f m e t a l s , " Svarocznoe Proizvodstro, No. 12, I (1961). 7. OJ<ADA, M. et al., "'Joint m e c h a n i s m o f u l t r a s o n i c welding," Tra,sactions c~['the Japanese Institute 0/' Metals, 4, 250 ( A u g u s t 1963). 8. BOWDEN, F. P. a n d TABOR, D., " T h e friction a n d l u b r i c a t i o n o f solids," Part 1, O x f o r d (1953). 9. BOWDEN, F. P. a n d TABor, D., "'The friction a n d l u b r i c a t i o n o f solids," Part 2, Oxford (1964).

Add a comment

Related presentations

Related pages

Ultrasonic welding - Wikipedia, the free encyclopedia

Ultrasonic welding is an industrial technique whereby high-frequency ultrasonic acoustic vibrations are locally applied to workpieces being held together ...
Read more


Die Domain "www.ultrasonic-welding.org" ist nicht verfügbar. Die Domain "www.ultrasonic-welding.org" ist nicht verfügbar.
Read more

Herrmann Ultrasonics: Technology leader in ultrasonic welding

Ultrasonic technology from Herrmann Ultrasonics for welding, sealing and laminating of plastics, packaging and nonwoven materials.
Read more

Ultrasonic Welding – Principles & Plastic Welders - Dukane

Learn about ultrasonic welding, including the process and principles as well as the plastic welder equipment available through Dukane Intelligent Assemble ...
Read more

Sonobond Ultrasonics Inc - Ultrasonic Welding--Metal ...

Sonobond has manufactured ultrasonic welders since 1960, including machinery for Metals, Nonwovens, Textiles, Filters, Batteries, body armor & Photovoltaic ...
Read more

How Ultrasonic Welding Works | HowStuffWorks

Ultrasonic welding uses high pitched sounds to bond materials together. Learn how ultrasonic welding works.
Read more

What is ultrasonic welding? – Herrmann Ultraschall

Ultrasonic welding of thermoplastic materials is a weld technology utilizing mechanical vibrations to generate heat due to molecular friction.
Read more

Sonobond Ultrasonics Metal Welding Technology | Sonobond ...

Innovative ultrasonic metal welders have been developed by Sonobond since they received the first ultrasonic metal welding patent in 1960. Oxidized and ...
Read more

Ultrasonic Welding — SONOTRONIC - Mit Erfolg verbunden ...

Ultrasonic welding is found everywhere where thermoplastics are used and strict demands are placed on the method of joining. Compared with other welding ...
Read more

Ultrasonic welding of plastics by Herrmann Ultraschall

With ultrasonic welding of plastics, plastic molecules are activated to create new cross-links among molecules. This creates a high-strength joint.
Read more