MD+DI Online is part of the Informa Markets Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Choosing the Right Welding Process

Article- Choosing the Right Welding Process

Medical Device & Diagnostic Industry Magazine MDDI Article Index   Originally Published MDDI September 2005 Manufacturing Choosing the Right Welding Process

Medical Device & Diagnostic Industry Magazine
MDDI Article Index

Originally Published MDDI September 2005


Plastics welding may not be the obvious choice, but it might be the best one for building your medical device.

Mike Johnston

The product is beginning to take shape on the computer screen. The rendered 3-D model rotates on various axes as artificial light and shadows play across the surfaces. Function is pretty well set, but form is still a bit of a question. “It sure would be nice,” you think, “if this could go together without adhesives or fasteners.”

The good news is that a great many medical devices can be joined without consumable adhesives or fasteners. Although snap- and press-fit applications are the least expensive options, they are often not permanent or secure enough. When that is the case, plastics welding is often the best solution.

When confronted with the variety of joining methods available, many designers face another choice: gain a lot of education in a short period of time, or go with an already-familiar process. Sometimes, however, the second choice leads to devices being assembled using a process that may not be the best one for the job.

It is important to understand and compare different methods in order to determine which may be best for a given application. Six popular plastics welding methods are presented here. The strengths and weaknesses of each process are described to help designers understand and compare techniques.

Ultrasonic Welding

Materials must be thin or stiff to be welded ultrasonically.

The ultrasonic welding process is used quite commonly for medical devices because its strengths nicely complement the requirements of medical device assembly. Economics often favor ultrasonic welding because cycle times are typically very short. In addition, the equipment and tooling costs are low in comparison with other processes. Ultrasonic welding will generally produce some flash and

particulate, but the flash is usually controllable and the particulate amount is usually small. For this reason, the process is considered to be cleanroom friendly. The process is relatively easy to automate. It is a popular method and is generally well understood by designers. The level of process control sophistication in ultrasonic welding is relatively high, and the process produces repeatable results if the process inputs are consistent.

The ultrasonic process, however, is constrained by acoustic limitations in the tooling design, which impose part configuration and contour limitations. Sound transmission through the material is quite important, and a near-planar joint is usually required so that the sound will travel the same distance to all points on the joint. Unless sections are very thin, material needs to be relatively stiff.

One of the more-serious limitations of this method is the vibration imposed on components by the process. Some materials, such as unreinforced semicrystalline materials, pose special challenges. Parts must be able to withstand the vibration, and the properties of these materials may be adversely affected by exposure to ultrasound.

Thermal Staking, Swaging, and Inserting

Thermal staking is a process that is easy to automate.

The thermal staking, swaging, and inserting process produces no particulate and does not introduce vibration to the components. It is fairly easy to automate and has relatively low tooling and capital costs. Staking or inserting can occur simultaneously on multiple planes and at multiple points, with no acoustic limitations on tool design. The process is fundamentally simple, and problems are relatively easy to troubleshoot.

Some materials that have a narrow melt range, however, are not recommended for the process, and the presence of heat can affect some assemblies. Because of the narrow melt range, it is easy to scorch some materials. They can also become very runny. The process can work from multiple directions when some special techniques are used. Lack of rigidity and high crystallinity can also cause challenges.

Heaters and thermocouples are consumable, and heater burnout detection can be expensive. The cooling air used during the process may also not be acceptable in some cleanrooms. The somewhat longer cycle time compared with ultrasonic welding may also be a limitation. Many medical device manufacturers do not like the more open-loop aspects of the process. The fillers in reinforced materials can pose an extra challenge.

Spin Welding

Spin welding ensures process control and creates a visible process.

Spin welding is a fast and simple process that produces hermetic seals with relative ease. Like the two previous methods, it is easy to automate and has relatively low tooling and capital costs. It is a very repeatable process with consistent process inputs. Modern spin welders are quite sophisticated and add the extra flexibility of many process variables. The process produces some flash that is usually controllable (more so than with ultrasonic welding). The process is relatively easy to understand and is also relatively straightforward and easy to debug. The spin welding process has no consumables and offers fast cycle times and hermetic seals. The sophisticated equipment does an excellent job of ensuring process control and creates a visible process.

The biggest limitation of spin welding is that the joint must be a circle geometry. Small parts require high rotation speeds (more than 10,000 rpm), and balance can be an issue if the spinning part is significantly lopsided. This method requires a lot of breaking torque in the servomotor.

One of the biggest limitations for medical device manufacture is the high levels of particulate produced by the process. Spin welding can be sensitive to melt-index and melt-temperature differences in the plastics. In addition, reinforced materials, such as those with 25–30% glass, can pose extra challenges. In these cases, it is best to look at other processes, such as ultrasonic or vibration welding. High-lubricity materials will also be a challenge, because the process relies on high friction.

Hot-Plate Welding

Hot-plate welding is the best process for materials with different melt indexes and temperatures.

Hot-plate welding is a process that can handle parts of just about any size or geometry—within reason. It does not create particles or introduce vibration to the assembly. The tooling and capital costs are typically higher than those of the small-machine processes mentioned above but less than other big-machine processes. It uses no consumables, and systems cost about half the price of vibration welders. Hot-plate welding may be the only choice for soft or flexible materials. It is good for processing soft elastomers, such as syringe bulbs.

It is a simple process that is stable, and it is usually relatively easy to obtain a hermetic seal in most materials. The process is fairly easy to troubleshoot. It is the best process for dealing with melt-temperature and melt-index differences in materials. It is also best for high-lubricity materials such as PTFE. The process can handle joint curvatures in multiple directions.

Hot-plate welding has among the longest cycle times of any of the welding processes discussed in this article. Materials with narrow melt ranges are quite difficult to weld well. The process is essentially open-loop. The presence of heat can affect some assemblies, and the process always produces flash (which can be trapped). Charred material can be introduced into the joint, which can affect use of the process for liquid reservoirs in some applications.

High-temperature materials in general, and especially high-temperature semicrystalline materials, will pose significant challenges because antiadhesion coatings cannot be used. In addition, these materials have a sharp melt temperature, and the skin can be hard to break on resolidified materials. Reinforced materials also pose a challenge because molten plastic can get onto the hot plate. For such materials, vibration welding is a good alternative.

Heaters and thermocouples are consumable, and energy consumption is high. Heater burnout detection will add some cost to the system. Typically, if one thermocouple burns out, it puts extra stress on the other heaters, so burnout detection is essential. The process will also add heat and possibly smoke to cleanrooms. Plate maintenance can introduce burned particles to weld joints, and plastic can get stuck on hot plates. Unfortunately, there is no magic coating to eliminate this problem.

Vibration Welding

Vibration welding is good for large or warped parts, but it should not be used for high-lubricity materials.

Vibration welding is usually thought of as a large-part process, but it can be used with smaller parts, especially if more than one part is made per machine cycle. It is relatively easy to obtain a hermetic seal, even if the material has a narrow melt-temperature range, because it applies such a large amount of energy. It has no consumables. This process is excellent for high-temperature materials and for the often challenging semicrystalline materials, whether they are reinforced or not.

The cycle time for vibration welding is short considering that this is a large-machine process, and the process itself is not very sensitive to changes in resin chemistry. Vibration welders can handle considerable curvature in the across-vibration direction, and the process uses high clamp forces that make it easy to deal with warped parts. It is one of the best processes for welding large parts. For the most part, the process is simple and stable.

Vibration welding always produces particulate and flash, though flash can be trapped. Some assemblies might be unable to withstand the vibration, and the oft-required joint flange may be a styling issue. The process cannot handle long unsupported walls, such as those in a multichamber tank, especially in the across-vibration direction. This is not a good process for high-lubricity materials such as PTFE. Parts must be reasonably rigid (although fabric and filter applications are usually feasible) unless the sections are thin. This process is not recommended for elastomers.

Most vibration welders utilize a hydraulic unit that may be banned from the cleanroom, although midsize and smaller machines can be constructed with a pneumatic lift table. Capital costs are high (around $100,000), as are tooling costs (minimum $12,000–$15,000). A vibration welder may be the noisiest machine on the shop floor, although generally the noise is limited to 80 dBA. It is likely that vibration welding is not used more in medical device manufacturing simply because designers do not think about it as an option.

Laser Welding

Laser welding can be used for materials with fillers and additives because it is not sensitive to variations in resin chemistry.

Laser welding is a vibration-free, particulate-free, and, in most cases, flash-free process that is nearly silent. Certain joint-design and machine concepts give the designer considerable freedom in part configuration. This is a high-precision process that is extremely scalable from very small to very large parts. Depending on the joint-design strategy, laser welds may be the most reworkable of all of the welded parts created by the processes presented here.

Part warp can usually be accommodated, and the process has relatively low sensitivity to variations in resin chemistry and additives. However, resin chemistry and additives can make a great difference in the ability to get weld energy to the weld zone. For example, some color pigments and fillers, such as talc, are not transparent to light at all. Also, different natural resins, such as POM and PTFA (PFA flux) scatter light differently. Welding through any significant thickness requires a lot of laser power and can affect capital costs. Very fine process control is possible. Sophisticated systems can have a high degree of programmability and can utilize fixture recognition to drive the economical lot size very close to one part. The process is nearly independent of part-size considerations.

Laser welding still has relatively high capital costs, though in some instances, it may be less than the other large-machine processes. Faster cycle times come with higher capital costs. The process is sensitive to part fit-up, and the costs of special colorants or absorbers can be a factor in some applications. In addition, some materials need shielding gases for processing. Laser welders are often purpose-built, and so there are few standard models commercially available. Maintaining the optics and laser components is best left to professionals, and the process has some consumables costs.


Many medical devices can be constructed without consumable adhesives or fasteners. When a secure bond is needed, plastics welding is often the best solution—even if snap- and press-fit applications are the least-expensive options. Understanding different methods is critical to determining which may be best for a given application. These six welding methods are the most commonly used.

Choosing the best plastics joining method for a particular medical device does not have to be difficult. Analyzing part size, shape, material, and function will narrow the field, and one or two processes will emerge as leading contenders. Decisions should be made with a close eye on both up-front costs and the total cost of ownership of the system over the long haul. If the decision-making process is done correctly, a stable and trouble-free process should result.

Robust processes involve the right equipment, clear identification of the largest process window among the candidate processes, and employee training.

Mike Johnston is national sales and marketing manager for Dukane Corp. (St. Charles, IL).

Copyright ©2005 Medical Device & Diagnostic Industry

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.