Ultrasonic Welding

Legacy context

Kubota Research’s documented heritage in infrared plastic welding began with a specific problem: high-strength composites were prone to shattering and spalling upon impact. Early work with the P-Wave™ IR radiation system demonstrated that consolidating a protective layer onto a reinforced composite could improve impact absorption without the degradation caused by excessive heat. This principle—delivering targeted infrared energy to consolidate materials while preserving fiber integrity—became the foundation for subsequent advances, including the development of fiber-reinforced thermoplastic composites with partners like DuPont-Toray.

That same focus on controlled, non-contact heating carries forward into modern discussions of thermoplastic assembly. As manufacturers evaluate different methods for joining components, the comparison between infrared plastic welding and ultrasonic welding often arises. Both techniques offer distinct advantages depending on the material, joint design, and production volume. The legacy of precise thermal management in infrared processes informs how these technologies are positioned today, particularly where large or continuous weld seams are required. This transition from early composite consolidation to contemporary joining questions reflects a continuous engineering thread, where the goal remains achieving strong, reliable bonds without compromising the base material’s structural properties.

Ultrasonic welding of thermoplastic composites is a fusion process that uses high-frequency mechanical vibration to generate frictional heat at the joint interface [2]. For engineers evaluating this process, the governing numbers are not abstract specifications; they are practical targets derived from recent aerospace-scale trials. The most immediately useful figure is the power requirement. For a 16-ply quasi-isotropic (QI) adherend welded in a standard sequential lamination and consolidation (SLS) configuration, a low-amplitude, longer-duration process consumed about 250 watts on average [6]. In contrast, a high-amplitude, short-duration weld on the same material required significantly more power, though the exact value is not specified in the source [6]. This 250-watt baseline is your starting point for estimating energy input per weld, not a universal limit.

The second controlling number is weld speed. Ultrasonic welding is described as very fast, applying heating and pressure in a single operation [7]. While the evidence does not provide a specific weld time in seconds, the process is contrasted with slower fusion methods, implying cycle times on the order of seconds rather than minutes for small coupons [7]. The third set of numbers comes from mechanical performance testing. In a study of joint disassembly and re-assembly using PEEK adherends with a low-melt polyaryletherketone (LM-PAEK) interlayer, the maximum stress at room temperature (approximately 23°C) was 6.56 ± 1.07 MPa, with a displacement at failure of 1.09 ± 0.21 mm [5]. At elevated temperature (121°C, equivalent to 250°F), the maximum stress increased to 8.47 ± 1.19 MPa, with a displacement at failure of 1.09 ± 0.21 mm [5]. Note the counterintuitive increase in strength at elevated temperature; this is a material-specific result, not a general trend.

Interpreting the Power and Temperature Numbers

The 250-watt average power figure [6] is not a machine rating; it is an absorbed power measurement during a specific weld schedule. For your process design, use this number to estimate energy density. If you are welding a larger area, scale the power proportionally, but be aware that the evidence notes different configurations have differing power requirements [6]. The low-amplitude, longer-duration approach is more energy-efficient per weld, but it may not be suitable for all geometries. The high-amplitude, short-duration approach trades energy efficiency for speed, which may be necessary for production throughput. Do not assume that the 250-watt figure applies to your part; instead, treat it as a calibration point for your own power monitoring.

The temperature data [5] is more subtle. The test at 121°C (250°F) is an elevated temperature condition, not a processing temperature. The fact that the weld strength increased at 121°C suggests that the LM-PAEK interlayer is not degrading at that service temperature. For your joint design, this means that if your application sees service temperatures up to 121°C, the weld is not the weak link, provided you use a similar interlayer system. However, the displacement at failure remained constant at 1.09 ± 0.21 mm across both temperatures [5], indicating that the failure mode is likely brittle and does not become more ductile with heat. This is a critical constraint for applications requiring energy absorption.

Process Configuration and Tooling Constraints

The evidence describes two unconventional practices used to scale ultrasonic welding to large-area welds [2]. First, the sonotrode was placed on the convex side of a V-shaped tool edge (VTE) [1]. This is not a standard flat anvil configuration. Second, the welding parameters were adjusted by operators for each specimen, treating the welds as best-effort trials [1]. This is a warning: ultrasonic welding parameters are not transferable across geometries without empirical tuning. In one study, eight specimens were welded—two at one organization and six at another—and all required operator adjustment [1]. For your engineering practice, budget for a parameter development phase on every new joint geometry.

Tooling is a major constraint. The evidence from induction welding, which shares similar thermal management challenges, highlights two specific issues that apply to ultrasonic welding as well [3]. First, heat dissipation: the setup at one facility recirculated liquid coolant through the coil to draw heat out of the top surface and used plaster tooling to pull heat from the rest of the coupon [3]. For ultrasonic welding, you must manage the heat generated at the interface, not just the heat applied. Second, squeeze-out protection: a defect example is squeeze-out from a lap shear weld where the adherends are not properly tooled [3]. This means your fixture must contain the molten polymer laterally, or you will lose joint strength.

Energy Director and Surface Preparation

Conventional ultrasonic welding uses an energy director (ED) with molded pyramids or holes, often called a mesh [7]. However, the evidence explicitly notes that ultrasonic welding has been proposed with a flat ED, meaning a resin film with no additional preparation, and even with no ED at all [7]. One manufacturer, Agile Ultrasonics, started with no ED in their work for the TDEA project [7]. This is a significant departure from textbook practice. For your process, the choice of ED affects power requirements and weld quality. A flat ED or no ED reduces part cost and eliminates a molding step, but it requires more precise control of amplitude and pressure to achieve consistent heating. The evidence does not provide a quantitative comparison of weld strength between ED types, so you must validate this empirically for your material system.

Testing and Qualification Considerations

The mechanical testing protocol in the evidence is specific. Tests were conducted at room temperature (~23°C) and elevated temperature (121°C/250°F) [5]. The test methods included lap shear and a modified three-point bend to assess debonding at the tip of a flange [5]. The bond film material, thickness, and failure mode were all studied variables [5]. For your qualification plan, replicate these conditions. The standard deviation on the max stress values is approximately 15-18% of the mean [5], which is typical for fusion welds but should inform your safety factors. If you are designing to a specific load, use the lower bound of the confidence interval, not the mean.

The evidence also mentions nondestructive evaluations (NDE) as part of the subelement demonstration [8]. The specific NDE methods are not named in the provided text, but the inclusion of this step in the process flow indicates that ultrasonic welds in critical applications require post-weld inspection [8]. Do not rely solely on process parameters; verify each weld.

Scaling and Practical Limits

The evidence is clear that scaling ultrasonic welding to large areas is not a linear process. The two unconventional practices—sonotrode placement on the convex side of the VTE and operator-adjusted parameters—were adopted specifically to prepare for scaling [2]. The power requirement of 250 watts was for a 16-ply QI adherend [6], which is a relatively thick laminate. For thinner or thicker stacks, expect the power to change, but the evidence does not provide a scaling law. The finite element thermal model referenced in the literature [4] is a tool you can use to predict temperature distribution, but the model is not described in the provided evidence. For your own work, consider building a thermal model to extrapolate from coupon-scale to subelement-scale welds, but validate it against physical trials.

In summary, the key numbers to remember are: 250 watts average power for a standard 16-ply QI weld [6], a maximum stress of 6.56 MPa at room temperature and 8.47 MPa at 121°C for a PEEK/LM-PAEK joint [5], and a constant displacement at failure of approximately 1.09 mm across temperatures [5]. These numbers are not design allowables; they are experimental observations from specific configurations. Use them to set expectations for power budgeting, thermal management, and joint strength, but always verify with your own material and geometry.