DEVELOPMENT OF AN APPROACH TO DETERMINE MINIMUM AMPLITUDE REQUIRED FOR ULTRASONIC WELDING

Transcription

1 DEVELOPMENT OF AN APPROACH TO DETERMINE MINIMUM AMPLITUDE REQUIRED FOR ULTRASONIC WELDING Miranda Marcus, EWI, Columbus, OH Harris Cohn, Josh Drechsler, and Dan Grodek, The Ohio State University, Columbus, OH Alex Savitski, Dukane, St Charles, IL Abstract One of the most important variables for ultrasonic welding is amplitude. While recommended amplitude guidelines are available for many generic materials based on accumulated industry experience, use of specialized and newly developed materials is rapidly increasing. To that end, it is desired to have a method for experimentally determining the minimum amplitude required for a material. This work investigates one possible method to determine minimum required amplitude. Introduction Ultrasonic welding is one of the most common plastic welding processes used in the industry. The three most influential parameters for this process are amplitude, duration, and force. While weld duration and force are parameters that can generally be easily adjusted during process development, amplitude requirements must be considered during the design process. The amplitude provided to the weld depends on the frequency of the equipment, the specific transducer being used, the ratio of the booster, and the design of the ultrasonic sonotrode. A typical approach is to design the sonotrode with as much gain as possible, while maintaining a stress lower than the fatigue limit of the material. The amplitude can then be reduced if needed via a lower ratio booster or electronic adjustment through the ultrasonic generator. There are some drawbacks to this approach, however. First, performance of the ultrasonic stack is improved when a similar amount of gain is provided from the horn and the booster. Second, electronically reducing amplitude via the generator reduces overall power available for welding to the amplitude percentage set. For example, if the amplitude were set to 50%, only 50% of the maximum power rating of the generator would be available, such as 1200W when using a 2400W generator. Third, using a higher than necessary amplitude can often lead to unacceptable part damage. Ultrasonic vibrations may cause cracking at sharp radii or small cross-sectional area and can cause damage to films, filters, and membranes. Welding at lower amplitudes can minimize all these risks. For all these reasons, it is important to be able to define a minimum amplitude required to weld a material. While general amplitude guidelines have been available and used by equipment suppliers for decades, they have been based on general application experience and are limited to more common material grades, and do not account for fillers. These guidelines are usually also overly conservative, recommending excessive amplitude values. They are most often intended to provide just an initial guidance in the process optimization, which diminishes their usefulness for specific applications. Developing a method to determine amplitude required for any material is essential as more and more customer materials are developed and used. Literature Review The internal heat generation rate, Q, during ultrasonic welding can be defined as: Q = ½ωE ε² Where ω is the frequency, E is the loss modulus of the polymer being welded, and ε is the strain amplitude. Amplitude is therefore, very influential to heat generation at the joint, theoretically [1, 2]. Liu, et al. performed an experimental procedure using the Taguchi method to determine the relative impact of welding parameters on ultrasonic welding of an amorphous (Polystyrene) and semi-crystalline () material. They showed that the amplitude of vibration has the most influence on weld strength, far exceeding even the influence of weld duration. Further, they found that the amplitude of vibration is more critical to the semi-crystalline material than amorphous materials [3]. A study by Han, et al. demonstrated that lowering amplitude may actually improve the weld characteristics. They found that, as amplitude was lowered, the maximum heating location moved further into the welding zone for more even heat distribution in the weld [4]. SPE ANTEC Anaheim 2017 / 1684

2 Equipment The 30 khz Dukane ultrasonic iq welder (Fig. 1) used for this experiment was capable of controlling a wide number of variables due to its advanced software package coupled with its servo-driven weld-head. maximum amplitude with the 1.5 booster used in the standard orientation would be 33.58μm. Tensile testing was performed on an Instron machine (Fig. 2) using tensile test tooling designed specifically for use with the Dukane ISTeP parts. The parts were pulled at a speed of 4 mm/minute. Figure 2. Tensile test tooling. Materials Figure 1. Dukane iq servo ultrasonic welder and tooling The amplitude of the stack was measured using a displacement gauge, showing a peak-peak amplitude value of 15μm when using a reverse oriented 1.5 ratio booster at 100% amplitude. When welding the PP material, the 1.5 ratio booster was used in standard orientation. When welding the PC material, the booster needed to be reversed to identify the minimum amplitude. One amorphous and one semi-crystalline material were chosen to develop the minimum amplitude determination process. The amorphous material used was Polycarbonate (PC), a Sabic Lexan 121R grade. The semi-crystalline material chosen was (PP), a Chase Plastics Pryme PPC-100NB-20M grade. The Dukane ISTeP style standard test parts were used which can both be leak and pull tested and are available with a variety of joint designs. For this work, the 90 angle energy director was used for both materials. A drawing of this part is shown in figure 3. Additionally, the theoretical amplitude of the ultrasonic stack was calculated. The gain of the sonotrode can be determined by the ratio of the mass before vs after the nodal point. Since density is constant and the nodal point is the geometric center, this can be simplified to the ratio of the cross-sectional areas. The sonotrode used has a backstock diameter of 47.7mm and a front diameter of 34.3mm. Using these values, multiplied by a factor of 0.8 to account for the radius of transition across the nodal point, the gain of the sonotrode can be estimated to be The 30 khz transducer provides 13.4μm of amplitude. Coupled with a reverse oriented 1.5 ratio booster and a sonotrode with a gain of 1.55, the theoretical amplitude that would be achieved is 13.9μm. This is quite similar to the measured value. Extrapolating from the measured value of 15μm using a reversed 1.5 booster which would provide approximately 0.67 gain, the Figure 3: ISTeP part drawing, 90 energy director (mm) SPE ANTEC Anaheim 2017 / 1685

3 Experimental Procedure The test was developed using the Melt-Match feature on the Dukane servo-driven ultrasonic welder. This feature allows the sonotrode to make contact with the part and stop at a position associated with a set trigger force. While maintaining this position, ultrasonic vibrations are initiated while not applying any additional force or downward motion to the parts. Once a drop in force is measured, indicating that the polymer has begun melting, then the welder begins its downward motion. For these experiments, a drop in force of 5% was used to indicate that melting had occurred. This feature is particularly useful to this test as it allows a distinct measure of the time at which melting begins in the material. This measurement can be taken as the time difference between the point at which ultrasonic vibrations are initiated and the point at which a drop in force is registered, hereafter referred to as time to melt initiation. This time to melt initiation was plotted versus a variety of amplitudes for each material. Initially, the materials were welded in 10% increments of amplitude and manually pulled apart to check for signs of welding. Once an estimated minimum amplitude was identified, subsequent welds were made in 5% increments from just below the estimated minimum amplitude to 20% above it. For example, the PP material was found to start welding at 40% amplitude, so the subsequent trials were made at 35% to 60% amplitude in 5% increments. For these trials, five parts were welded at each amplitude setting and tensile tested. Onset of melting was identified both through a drop in force when exciting the material with ultrasonic vibrations with no vertical collapse, and through weld strength as measured via tensile testing. Results and Discussion The initial trials using 10% amplitude increments and manual pulling showed that requires a minimum peak-peak amplitude of about 13.4μm. Parts welded at 6.7μm and 10.1μm easily came apart under manual inspection. This finding is supported by the weld data results, which show that the higher amplitudes used results in a quicker force drop after initiation of ultrasonic vibration (Fig. 4). The drop in force indicates reduced resistance due to material melting of the parts. As no downward motion is being applied, this drop in force is solely due to application of ultrasonic vibrations. Figure 4: Force vs. Time at various Amplitudes and no vertical motion for Following these initial trials, five parts were welded at 5% amplitude intervals around the minimum amplitude as indicated by the initial results. These parts were tensile tested. Data was taken for one representtative sample from each ampltiude setting to determine the time to initiate melt in the material. The time for a 5% drop in force to occur and the average tensile force were plotted versus amplitude (Fig. 5). Figure 5: Amplitude vs. Time to Melt Initiation for The tensile testing results shows a simliar gradual increase in average weld strength as amplitude is increased (Fig. 6). The standard deviation for PP indicates a fairly reliable weld process across all amplitudes used. Figure 6: Amplitude vs. Tensile Strength for SPE ANTEC Anaheim 2017 / 1686

4 Polycarbonate The initial trials using 10% amplitude increments and manual pulling showed that Polycarbonate requires a minimum peak-peak amplitude of about 6μm. Parts welded at 3μm and 4.5μm easily came apart under manual inspection. Unlike the sharp drop in average time to melt, average tensile force gradually increases with increasing amplitude. However the wide standard deviations indicate an unreliable process for PC at this range of amplitudes. This finding is supported by the weld data results, which show that the higher amplitudes used results in a quicker force drop after initiation of ultrasonic vibration (Fig. 7). The drop in force indicates deformation of the parts. As no downward motion is being applied, this drop in force is solely due to application of ultrasonic vibrations. Figure 9: Amplitude vs. Tensile Strength for Polycarbonate It was noted during the welding process that many of the Polycarbonate parts showed bubbling in small areas around the joint, as shown in Figure 10. Figure 7: Force vs. Time at various Amplitudes and no vertical motion for Polycarbonate Following these initial trials, five parts were welded at 5% amplitude intervals around the minimum amplitude as indicated by the initial results. These parts were tensile tested. Data was taken for one representtative sample from each ampltiude setting to determine the time to initiate melt in the material. The time for a 5% drop in force to occur and the average tensile force were plotted versus amplitude (Fig. 8). The samples showing a time of 16 seconds represent no melting before reaching the set timeout signal which aborts the process. Figure 10: Bubbling of weld on Polycarbonate This bubbling was primarily noted at higher amplitudes and may indicate a problem with the quality and consistency of material used. One likely cause is moisture content in the parts. Cross-Section Evaluation Cross-Sections can be used to evaluate weld quality. One part was cross-sectioned and heat treated to determine if a poor weld could be discerned via this analysis method (Fig. 11). This cross-section was taken of a PP part welded using 16.8μm of amplitude. Figure 8: Amplitude vs. Time to Melt Initiation in Force for Polycarbonate SPE ANTEC Anaheim 2017 / 1687

5 this experiment show that weld strength continually improves with increasing amplitude. It would be worthwhile to investigate if an upper limit of amplitude can be achieved, above which weld strength becomes less consistent or decreases. Figure 11: Cross-Section of PP, as polished After polishing, the cross-sectioned part was heat treated by applying hot air to induce stress relief on the polished surface of the polymer. The resulting surface shows that a poor weld was made (Fig. 12). Additionally, it would be very worthwhile to conduct a similar experiment using other materials, for example a non-hygroscopic amorphous material such as Polystyrene and a harder semi-crystalline material such as Polyphenylene Sulfide. Acknowledgements A great deal of thanks goes to Dukane for providing the equipment and material required to do this research. Additionally, many thanks to Professor Avi Benatar whose expertise and support made this project a success. References Figure 12: Cross-Section of PP, after heat treating This result may indicate that, while melt can be generated at low amplitudes, to achieve a sufficient intermolecular bond to obtain an objectively good weld, higher amplitude is needed. Conclusions This work provides a framework for an experimental method to determine minimum required amplitude for any material. The strong correlation of time for the material to begin melting, as noted by a drop in force, to weld strength indicates that using the Melt-Match feature on the Dukane servo-driven welder to estimate minimum required amplitude is viable. 1. Grewell, Benatar, Park, Plastics and Composites Welding Handbook, Hanser Publishing, Levy, Le Corre, Poitou, Ultrasonic welding of thermoplastic composites: a numerical analysis at the mesoscopic scale relating processing parameters, flow of polymer and quality of adhesion, International Journal of Material Forming, Liu, Lin, Chang, Wu, Hung, Optimizing the Joint Strength of Ultrasonically Welded Thermoplastics, Advances in Polymer Technology, Han, Liu, Chong, Yan, Analysis of heat generation characteristics in ultrasonic welding of plastics under low amplitude conditions, Journal of Polymer Engineering, Despite the initial indication of a distinct minimum amplitude during the first trials, however, there does not appear to be a distinct minimum amplitude required for either material, although time to melting at low amplitudes is too long to be reasonable for most manufacturing processes. This is perhaps due to the materials selected. Polycarbonate is an amorphous polymer, which is expected to have a gradual shift from unweldability to weldable., while a semicrystalline material, is very soft and may be able to be pushed together to form a mechanical surface bond even with poor melting. Future Work This experiment directly leads to several possible interesting investigations. First, the amplitude chosen for SPE ANTEC Anaheim 2017 / 1688