A Study on Through-the-Thickness Heating in Continuous Ultrasonic Welding of Thermoplastic Composites

Abstract

Continuous ultrasonic welding is a promising technique for joining thermoplastic composites structures together. The aim of this study was to gain further insight into what causes higher through-the-thickness heating in continuous ultrasonic welding of thermoplastic composites as compared to the static process. Thermocouples were used to measure temperature evolutions at the welding interface and within the adherends. To understand the mechanisms causing the observed temperature behaviours, the results were compared to temperature measurements from an equivalent static welding process and to the predictions from a simplified heat transfer model. Despite the significantly higher temperatures measured at the welding interface for the continuous process, viscoelastic bulk heat generation and not thermal conduction from the interface was identified as the main cause of higher through-the-thickness heating in the top adherend. Interestingly the top adherend seemed to absorb most of the vibrational energy in the continuous process as opposed to a more balanced energy share between the top and bottom adherend in the static process. Finally, the higher temperatures at the welding interface in continuous ultrasonic welding were attributed to pre-heating of the energy director due to the vibrations being transmitted downstream of the sonotrode, to reduced squeeze-flow of energy director due to the larger adherend size, and to heat flux originating downstream as the welding process continues.

Keywords: CF/PPS; energy director; fusion bonding; heat transfer; high-frequency welding; joining.

Figure 1

Schematics of static (a) and continuous (b) ultrasonic welding processes for thermoplastic composites.

Figure 2

Cross-sectional micrographs of a (a) continuous (adapted from [15]) and a (b) static ultrasonic weld. Welding force 500 N, vibration amplitude 80 $$\mathsf{\mu }$$m, welding speed 35 mm/s (a), equivalent vibration time 430 ms (b) [14]. The same clamping jig and clamping configuration is used in both cases. The red circles indicate squeeze out of fibres and/or resin.

Figure 3

(a) Custom-built welding machine for static and continuous ultrasonic welding, (b) close-up of the continuous welding set-up visualising relative sonotrode and consolidator placement, and (c) schematic side-view of clamping distance and sonotrode placement.

Figure 4

Schematic side and top view of the temperature measurement configurations used in this study on 220 mm wide and 15 mm wide adherends.

Figure 5

Schematic model with boundary conditions for heat transfer model of static welding setup.

Figure 6

Temperature evolution for the continuous welding process at the weld interface for: (a) reference case, (b) higher welding speed case (65 mm/s), (c) lower amplitude case (70 $$\mathsf{\mu }$$m), and (d) damping case. TC1 to TC5 were respectively located under the sonotrode during the five grey areas. The red dashed line indicates the melting temperature of PPS ($$T_m$$, 280 $${}^{\circ }$$C) as experimentally determined by DSC analysis.

Figure 7

Interface temperature (black curves) and measured downward vertical displacement (grey curves) of the sonotrode for static welds on (a) 15 mm wide and (b) 220 mm wide adherends. It should be noted that one of the two welds in (b) was made 40 mm to the left from the intended location in Figure 4a. The red dashed line indicates the melting temperature of PPS ($$T_m$$, 280 $${}^{\circ }$$C) as experimentally determined by DSC analysis.

Figure 7

Interface temperature (black curves) and measured downward vertical displacement (grey curves) of the sonotrode for static welds on (a) 15 mm wide and (b) 220 mm wide adherends. It should be noted that one of the two welds in (b) was made 40 mm to the left from the intended location in Figure 4a. The red dashed line indicates the melting temperature of PPS ($$T_m$$, 280 $${}^{\circ }$$C) as experimentally determined by DSC analysis.

Figure 8

Temperature evolution for the continuous welding process at the weld interface and through the thickness for (a) the top adherend and (b) the bottom adherend. The grey areas indicate the time span during which a specific thermocouple was located under the sonotrode. The red dashed line indicates the melting temperature of PPS ($$T_m$$, 280 $${}^{\circ }$$C) as experimentally determined by DSC analysis.

Figure 9

Temperature evolution for the static welding process at the weld interface (blue curves) and through the thickness (black curves) for (a) the top adherend and (b) the bottom adherend. The temperature evolution at the weld interface in (a) is the same as Figure 7a. The red dashed line indicates the melting temperature of PPS ($$T_m$$, 280 $${}^{\circ }$$C) as experimentally determined by DSC analysis.

Figure 10

Modelled through-the-thickness temperature evolution due to heat transfer (HT) as a result from experimental (exp) temperature input at the weld interface for (a) the continuous and (b) static temperature evolution together with a representative experimental temperature evolution as a reference. The grey area (a) and the vertical line (b) indicate the time span during which the thermocouples were located under the sonotrode. Note that in (a) no experimental temperature evolution for the top adherend is shown as it was deemed less trustworthy due to severe overheating.

Figure 11

Representative cross-sectional micrographs of continuous ultrasonic welds for (a) higher speed case (65 mm/s), and (b) lower amplitude case (70 $$\mathsf{\mu }$$m). Red circles indicate squeeze-out location.

Figure 12

Representative fracture surfaces from continuous welds for (a) the reference case, (b) higher welding speed case (65 mm/s), and (c) the lower amplitude case. Red arrows indicate voids, white arrow indicates unwelded area together with voids, and black arrow indicates area without connection between top and bottom adherends.

Figure 13

Combined interface temperature evolutions of continuous (reference case) (Figure 6a) and static (Figure 9a) processes superimposed for time that thermocouple experiences vibrations directly under sonotrode. The red dashed line indicates the melting temperature of PPS as experimentally determined by DSC analysis.

Figure 14

Combined interface temperature evolutions of the continuous ultrasonic welding process for reference case (Figure 6a) and damping case with the consolidator as damping unit (Figure 6d without TC4) superimposed for time that thermocouple experiences vibrations directly under sonotrode (grey area). The red dashed line indicates the melting temperature of PPS as experimentally determined by DSC analysis.

Figure 15

Combined interface temperature evolutions of the reference case (Figure 6a) of the continuous process, the static process (Figure 9a), and the temperature evolutions for (a) the higher welding speed case (65 mm/s) (Figure 6b), and (b) the lower amplitude case (70 $$\mathsf{\mu }$$m) (Figure 6c). The temperature evolutions were superimposed at the moment they were under the sonotrode for the first time. The red dashed line indicates the melting temperature of PPS as experimentally determined by DSC analysis.