Joint Design for Ultrasonic Welding
These tips will help you design plastic parts for optimal assembly with ultrasonics
When designing plastic parts to be assembled by ultrasonic welding, engineers have several options: a butt joint, a step joint, a tongue-and-groove joint, and a shear joint. Which to choose depends on many factors, including the materials; the size and rigidity of the parts; and performance requirements, such as joint strength, cosmetics and leak-tightness.
“It all depends on what the end-product needs to be,” says Stephen Potpan, product manager for Rinco Ultrasonics USA. “Does it need to be hermetically sealed? Or does it just need to be strong and not come apart?
“Part size also comes into play,” he continues. “For example, a shear joint is the recommended design for a hermetic seal, but the part geometry may not allow for it. The design must provide enough room to incorporate certain features.”
The butt joint with an energy director is one the most common joint designs for ultrasonic welding, and it’s the easiest to mold into a part. An energy director—a small, triangular ridge molded into one of the mating surfaces—is essential for this design. The ridge can be molded in the bottom part (pointing up) or the top part (pointing down).
The energy director limits initial contact to a small area and focuses the ultrasonic energy at the apex of the triangle. During the welding cycle, the concentrated ultrasonic energy causes the ridge to melt and the plastic to flow throughout the joint area, which helps bond the parts together.
“An energy director helps to initiate the melting process,” says Cristhian Mayorga, associate applications engineer with Emerson Industrial Automation.
Without an energy director, the ultrasonic energy will not be evenly distributed along the joint line, which means more time and energy will be required to produce the same result. That extra time and energy can translate into unacceptable flash or tooling marks on the surface of the part.
The point of the triangle can be molded at a 90-degree angle or a 60-degree angle. The former is best for amorphous resins, such as ABS or polystyrene. The latter is better for semicrystalline materials, such as nylon and polyethylene. As a rule of thumb, the width of the energy director should be no more than 20 percent to 25 percent of the wall thickness. The height of the energy director can range from 0.01 to 0.04 inch, depending on the material’s melt temperature.
A step joint with an energy director is relatively easy to implement in an injection-molding tool. This joint is usually much stronger than a butt joint, since material flows into the clearance necessary for a slip fit, establishing a seal that provides strength in shear as well as tension.
“My personal preference is a step joint with an energy director,” says Brian Gourley, welding group sales manager at Sonics and Materials Inc. “From a molding standpoint, this part design gives you a little more leeway in terms of tolerances for a dimensional fit. With a tongue-and-groove joint, the tolerances have to be a little bit tighter so the mating surfaces can fit together without having side-wall interference.”
“Another favorite is the shear joint,” adds Gourley, an engineer with 35 years of experience in plastics assembly. “One can be better than the other. It depends on the application. In some cases, if all you need is a mechanical weld, a step joint is probably acceptable. But, if the requirements for the application are stricter, the shear joint might be better than the step joint, especially if you need a hermetic seal or the assembly needs to hold pressure.”
The step joint supports self-centering of components, and it’s recommended when a good cosmetic appearance is required.
“A step joint will protect against flash, at least from one direction,” explains Jason Barton, director business development for the Americas at Dukane. “Most customers don’t want to see flash on the outside of the assembly, but they don’t mind if it goes into the inside. On the other hand, if it’s a vessel that’s involved with filtration, then the customer may not want flash inside or out. That’s where a tongue-and-groove joint comes in.”
Joint designs with energy directors tend to work best with amorphous materials—plastics with good stiffness and low to medium melt temperatures.
“Materials such as polyolefin, polyethylene and polypropylene don’t always work well with an energy director,” says Gourley. “Those materials are soft and compressible. They don’t transmit ultrasonic vibrations efficiently, so a step joint with an energy director is typically not our first choice. We’ll look at something that’s going to be a little bit more aggressive, such as a shear joint.”
The greatest strength is usually attained with a tongue-and-groove joint. Gap dimensions with very small clearances create a capillary effect that causes the melted plastic to penetrate through the entire joint area.
Like the step joint, this design facilitates self-location of the parts, and it hides flash both internally and externally. It also protects the weld line from ambient air flow, which can sometimes adversely affect the welding process.
“A tongue-and-groove design is best for a hermetic seal, but if the part thickness can’t accommodate that, then we’ll use a shear joint,” says Mayorga. “A shear joint can be done with a thinner wall thickness, but it will require more amplitude to weld the parts.”
The tongue-and-groove joint design requires relatively thick walls. Whereas the minimum wall thickness for a step joint is 0.062 inch, the minimum wall thickness for a tongue-and-groove joint is 0.09 inch. The tongue will feature an energy director.
The shear joint has proved to be successful for welding semicrystalline plastics. With large joining distances, this joint design typically produces air-tight and high-strength welds.
“If a major criteria is leak-tightness, the shear joint is a good choice, depending on the material and the geometry,” says Barton. “If you’re trying to weld a vessel made out of nylon, you want a shear joint.”
Shear joint welds are achieved by first melting the contacting edges, then continuing the melt along the vertical walls as the parts telescope together. Telescoping prevents exposing the weld region to air, which could cool it too rapidly, causing brittleness. Weld strength is determined by how far the top part telescopes into the bottom part. A weld depth of 1.25 to 1.5 times the wall thickness of the part will create a weld that is nearly as strong as the surrounding wall.
The parts are guided together by a lead-in, and the weld is controlled by the amount of interference between the two parts. Rigid side-wall support is important with shear joint welding to prevent part deflection during welding.
“A shear joint will require more work of the ultrasonic welder to overcome the interference fit between the two parts,” notes Gourley. “As a result, the parts will be exposed to the ultrasonic vibrations longer than they would be with a step joint, so there is a possibility of tooling blemishes on the parts.”
Where flash is objectionable, flash containment traps should be included in the design.
“Flash traps are tricky, though,” cautions Gourley. “That plastic material flowing outside of the weld joint can actually aid in weld strength. The flash that escapes from the weld joint will travel between the two parts. But, if there’s a flash trap, that material flows into the trap. You’ll get a cosmetic joint, but you may not get the cross-sectional strength from the additional material that is bonding the two pieces together. It’s a fine line.”
Big Picture Design
Outside of the joint itself, there are some general considerations engineers should think about if they intend to assemble a product via ultrasonic welding.
One is wall thickness. “Thin walls may not have enough stiffness to transmit vibrations to the joint area to create an effective weld,” Gourley points out. “If the wall is too thin, maybe 0.06 inch or less, it’s going to act like a spring. It will move in a lateral direction instead of a vertical direction, and it will rob energy from the joint area.”
Another consideration is the overall size of the assembly. “There are limitations in terms of a single horn contacting a single part,” says Barton. “For ultrasonic welding, the part typically has to fit inside a 10-inch by 10-inch box or maybe a 12-inch by 12-inch box. But that depends on the materials, too. You’re not going to weld a 10 by 10 nylon box, but you might be able to weld a 10 by 10 polystyrene box.”
And, because the horn must have intimate contact with the part, engineers should avoid features that could prevent the horn from reaching its target. Machining excessive pockets into the face of the horn could impair its ability to vibrate.
Designs with straight lines or curves can be welded, but rounded corners are preferred over sharp ones. “Ultrasonic energy tends to migrate to sharp areas and cause cracks,” notes Potpan.
Near-field welds—in which the horn is 0.25 inch or less from the joint—are preferred over far-field welds, where the horn is more than 0.25 inch from the joint. The latter will require higher amplitudes and longer weld times than the former.
The Early Bird
While ultrasonics professionals might have their personal preferences for joint designs, there’s one piece of advice they all agree on: Get resin and welder suppliers involved early in the design stage. Many suppliers offer design guidelines and worksheets to help engineers get started when designing a part for ultrasonic welding.
“The biggest issue we see is that the parts have already been designed and are off to the mold maker,” says Potpan. “Then, when we get involved and mold changes are needed, it becomes very expensive.
“We always try to get with the engineers when they are doing the design work in CAD. That way, when we recommend changes to the joint design, we are just changing a computer model not a steel mold.
Gourley agrees. “Get us involved early in the design stage,” he advises. “If we can review a CAD file and have a conversation about design, we can help engineers save time, money and anguish. It’s a lot easier to make changes on a CAD file than it is to change a mold.”
Joint Design Guidelines: One on One With Herrmann Ultrasonics
Ultrasonic welding is arguably the most popular method for assembling plastic parts. The technology is fast and economical. It requires no consumables, and it can be applied to a wide variety of applications. Ultrasonic welding can create strong, hermetic welds that are visually appealing.
However, various design considerations must be taken into account to ensure optimum results. Recently, we sat down with Michael Lindsay, an application engineer with Herrmann Ultrasonics Inc. to discuss those considerations. For more information on ultrasonic plastic welding, call Herrmann Ultrasonics at 630-626-1626 or visit www.herrmannultrasonics.com.
ASSEMBLY: There are several joint designs for ultrasonic welding—step, tongue and groove, shear, etc. Is one better than another?
Lindsay: There is no single weld joint that is universally considered the “best” for all applications. They all have their own unique pros and cons, which have to be taken into consideration based on the requirements and design constraints of each application.
ASSEMBLY: I imagine strength is the No. 1 criteria for a weld, but what role do other factors, such as aesthetics or leak-tightness, play in joint design?
Lindsay: All of those factors will certainly influence what the ideal joint design should be. Each weld joint has features that might lend themselves to certain outcomes. For example, a simple step joint works well for most applications where only strength and aesthetics are needed. If you require a leak-tight seal, then a mash joint, which is an improved version of the troublesome shear joint, would be better. Now, if you are welding a medical device that has internal fluid pathways, then a tongue-and-groove joint will isolate internal flash and prevent it from breaking loose into the fluid path.
As we discuss an application, we look at all of the requirements, the overall part design, wall thickness, the material, and limiting constraints, then choose the best option that fits.
ASSEMBLY: What role do the materials play in joint design for ultrasonic welding? Does a particular material rule out a particular design?
Lindsay: Most materials would be fair game for any given weld joint under certain circumstances. However, there may be other mitigating factors in combination with a material choice that could influence certain decisions. For example, when using softer semicrystalline materials, such as polypropylene, on larger parts, the walls can become very flexible and unstable. That might steer us away from recommending a step joint, because there could be cases where the walls warp and flex inward, resulting in internal joint misalignment. If you add a stiffener, such as glass, to the plastic, then that might offset that issue to some degree.
ASSEMBLY: Can materials be mixed?
Lindsay: As a rule of thumb, to achieve a true molecular bond, you need to weld the same material to itself. However, there are some exceptions where material properties are close enough that they blend sufficiently when welded. For example, ABS can be welded to PMMA or PC with very good results. There are a small handful of other material combinations that have varying degrees of compatibility, and we must consider the overall application requirements to see if the limitations are acceptable.
In cases of staking, or embedding of membranes, then you can break this rule more readily. With staking, for example, a post is mashed down and re-formed like a rivet, so you could stake a metal plate to a plastic housing, for example. We also work with embedding thin pressure-balancing membranes and filter materials directly to injection-molded surfaces. In this case, you are melting the plastic surface material, which flows into the porous structure of the membrane to capture it mechanically. This is analogous to mashing a piece of chewing gum into the carpet. It’s mechanically captured, but not a true molecular bond.
ASSEMBLY: What about part dimensions? Are there minimums or maximums in terms of wall thickness? What about overall part size (length, width, height)? Any rules of thumb?
Lindsay: All of the common weld joints have specific dimensional and clearance recommendations. You must allow for the parts to move freely so that they are not “pressed” together during pre-assembly. The internal vertical gaps should be large enough to contain melt volume, yet not too large that side-to-side play would allow joint misalignment.
Certain joint types require more wall thickness to properly incorporate, as well. For example, the tongue-and-groove joint requires a minimum wall thickness of 3 millimeters. If you scale that to a really small part, then that’s not always practical, so you may need to look at an alternate joint.
You want to look at the overall part design and try to make sure the sonotrode (horn) contact surface has direct line of site to the weld joint through the Z axis whenever possible. Conversely, you want to make sure you have fixture support directly underneath the weld. The joint should allow for sufficient part travel during the weld, avoiding internal hard-stops that limit collapse.
When considering overall part size, it’s sort of a loaded question. If you are welding a lid with a continuous perimeter weld to a sealed container, for example, then the size of the part is bound by limitations of how big a horn can be made. However, if you are spot-welding individual points on a car door panel for example, then you can jump from point to point and perform multiple hits with a single sonotrode, and then the sky is the limit.
As a general rule for a single continuous weld, we use “shoebox size” as a generic estimate for the maximum size for which a single block-style sonotrode can be made.
ASSEMBLY: Do you like to see straight lines or curves? Round corners or sharp ones? Rigid parts or flexible? Does it matter?
Lindsay: Square and round objects can both be welded successfully, but each has unique challenges. With square parts that require a liquid-tight seal, you want to avoid sharp, 90-degree directional changes as the joint turns the corners. Material tends to flow unpredictably in those areas and can result in leak paths. We recommend to round off the internal joint so it sweeps around the corners for that reason.
This also goes for internal support walls, stiffening ribs or alignment post features. Sharp, inside 90-degree corners tend to accumulate stress under vibration loads, so you want to incorporate a fillet radius on those features to prevent them from breaking loose.
With round parts that have a dome-shape cover, such as 3D handles, you might not always be able to achieve sonotrode contact directly above the weld joint, especially if the shape is such that the tangent point of the dome is inside of the plane where the weld joint resides. We usually don’t sink the part into the sonotrode pocket past the tangent point to avoid scrubbing, so in those cases we need to make sure a stiffer material is used to compensate for the potential loss of energy. Overall, stiffer materials weld more efficiently than flexible ones.
ASSEMBLY: When looking at a part, is there anything about it that might make you think right away, “this is going to be easy to weld” or “this is going to be hard”?
Lindsay: Yes, frequently see what we call “core applications” that are similar to other welds that we are experienced with, and can usually determine pretty quickly whether it will be a relatively “text-book” application or a challenging one. In cases where we see something outside of the norm, we might determine, based on size, shape, material selection, etc, that it’s going to be a bigger challenge. Often we can discuss the “red flags” with the customer and make recommendations that would bring the application more into the feasible realm. In some cases, however, where we feel the risk is too high even with changes, we have to make a difficult decision and steer the customer to a different joining technology.
ASSEMBLY: What’s an energy director? Why is it necessary? What should it look like?
Lindsay: An energy director is a feature that concentrates the contact point and directs the energy to a pinpoint location. This is important for the initiation of the weld.
If you are welding two surfaces flat-to-flat, that means you have a larger surface area at initial contact to heat up and melt. This requires more time and energy, and thus extended exposure to the process. In addition, the longer exposure and larger surface area causes the heat to migrate further up into the walls away from the weld joint.
An energy director concentrates the vibration in a smaller surface area, so it can reach melt temperature much quicker. This shortens the weld time, lowers the overall energy required, and helps keep the melt contained in a more localized area around the weld joint. A good energy director will typically be a triangular shape with a 60- to 90-degree point. It will be approximately 0.3 to 0.7 millimeter tall, scaled to the size of the part. You want to make the point as sharp as possible, which usually means that’s an EDM feature in your mold. Flatness along the length should be consistent as well.
ASSEMBLY: What’s the difference between a far-field weld and a near-field weld?
Lindsay: A far-field weld is where the sonotrode contact surface is further away from the actual weld joint, so the vibration energy has to travel through a longer distance to reach the joint. Near-field means the sonotrode is closer. While there are plenty of examples of successful far-field applications, near-field is preferred, especially when you are using softer, semicrystalline thermoplastics, which tend to dissipate the vibration energy as it travels through the walls of the part.
Sometimes we may look at an application and make a decision to weld in the far-field orientation, especially if the near-field contact surface has features that prevent good sonotrode contact. Of course other mitigating circumstances have to be considered while making that decision.
ASSEMBLY: What’s the most common mistake engineers make in terms of joint design for ultrasonic welding?
Lindsay: Generally speaking, a common issue we see is going too far down the wrong design path before engaging in discussions about weld feasibility. When the final assembly process is considered as an afterthought, we sometimes have to suggest last-minute, dramatic changes to the overall part design that may not be well-received. However, in cases where a weld joint has been considered ahead of time, I would say the most common areas that need attention are with energy directors being incorrectly sized or positioned.
ASSEMBLY: Have you ever been presented with a design that could not be welded with ultrasonics? What was it? How did you resolve the issue?
Lindsay: Yes, it does happen. The most common reason we run into non-feasible applications is probably due to part size. When evaluating an application for feasibility, we have to consider the limitations for how large a sonotrode can be made. If an assembly requires a continuous weld, for example, like a sealed container, and it is larger than the maximum sonotrode size, then that would clearly be a non-feasible application. The most likely work-around would be to reduce the size of the welded cover to within the feasible range of the tooling.
Another example of a difficult application would be a multistage weld, where you are trying to stack multiple components in different planes within the same part (like a sandwich), and weld them all at the same time. For these applications, we always recommend to weld each subassembly individually, then combine the subassemblies for one final overall assembly weld.
ASSEMBLY: How much, if at all, can an ultrasonic welding system overcome a suboptimal joint design? Can you still weld a suboptimal design by adjusting welding parameters (power, amplitude, etc.) or better tooling? Or is it better to redesign the parts?
Lindsay: It is not uncommon to see suboptimal applications, and to some degree, you can overcome minor issues through manipulation of weld parameters or tooling design. In extreme cases, you might end up with an exotic weld process that can result in a very narrow process window. This leaves little to no margin for error with tolerance stack-ups. Additionally, the “fix” for one issue sometimes can cause a new problem.
We also realize that when working with a customer on a design, there is always a potential for conflicting motivations between the different groups within that organization. The design team wants certain features, the manufacturing team wants it to be easy to build, the marketing team wants it to look and feel a certain way, and the accounting team wants to make it as inexpensive as possible. It is our job to make sure that everybody is on the same page with regards to the weld design. If exceptions are to be made, we need to make sure the customer is aware of the potential risks and the impact they might have ahead of time.
ASSEMBLY: What other advice would you give to engineers when design plastic parts to be jointed with ultrasonic welding?
Lindsay: It is never too early to get us involved in the discussion about ultrasonic welding on a new application. The earlier in the design phase we can give our input, the easier it is to incorporate the correct features. Also try to follow the guidelines and “best practices” of material selection, joint features, etc., as best as possible. Lastly, quality molded parts are very important. We sometimes see designs that look great on paper, then the parts come in from the molder with issues that may be problematic for welding, such as tolerance deviations, sink marks, warping, parting lines or gates in bad locations. It’s very important to avoid these issues in the molding process to ensure the highest chance of welding success.