Subscribe to our newsletter to receive the latest news and events from TWI:

Subscribe >
Skip to content

Friction welding of plastics

Fig.1. An air intake manifold joined by friction welding
Fig.1. An air intake manifold joined by friction welding

Friction welding of thermoplastics is a long established technique usually employed for joining injection-moulded parts. The welding process has found many applications ranging from automotive, for example air intake manifolds (see Fig.1) and expansion tanks, through to domestic appliance components such as a cistern ball float. Experimental applications of friction welding for thermoplastics have included welding polyethylene pipes for gas and water distribution.

There are six identifiable variations of the friction welding process; linear, orbital, multi-directional, rotational, angular and friction stir.

The linear and rotational forms of the friction welding process are used extensively in industrial applications. Figure 2 shows a typical linear friction-welding machine. Welding machines using orbital and multi-directional techniques have only become available in recent years. The angular friction welding process has only been used in a limited number of commercial applications and equipment is not commercially available.

The friction stir process is still under development for plastics, although now used extensively for metals.

Fig.2. An example of a typical linear friction welding machine
Fig.2. An example of a typical linear friction welding machine

Thermoplastic friction welding processes

Linear friction welding (also known as vibration welding) of thermoplastics involves rubbing together, under axial force, one component in a linear reciprocating motion against a fixed stationary component. The frequency of the vibration is typically between 100 and 240Hz with a peak-to-peak vibration movement of 1 to 4mm.

Rotational friction welding (or spin welding) includes rotating one part in a continuous circular motion against another part, under axial force. The typical rotation speed is between 1200 and 3500rpm.

Orbital welding involves rubbing together the thermoplastic parts, under axial force, in an orbital motion at the interface. Similar to linear friction welding, the frequency of operation is around 200Hz with an off-axis deflection between 1 and 2mm. The orbital motion has been adapted on some equipment to give a multi-directional, non-uniform vibration pattern.

Angular friction welding, is designed to allow circular components to be welded in a vibration mode. The components are rubbed together in a reciprocating motion, through a few degrees (typically 2 to 5°), during the welding process giving an arc of vibration motion at the component interfaces.

The final friction process, friction stir, involves a moving non-consumable tool that is forced between the parts to be welded, which are held fixed. The tool can be either a rotating pin or a blade that is vibrated in a linear reciprocating motion, either in line with the joint or perpendicular to it.

Process operation

In all the thermoplastic friction welding processes, the heat generated by the rubbing action must be sufficient to melt and flow the plastic at the weld interface. Sufficient heat is generated by a combination of weld time, weldforce and interface velocity, determined by either the reciprocating or rotational motion.

Figure 3 shows a schematic of the material displacement at the weld interface during the welding cycle. In all friction processes, except friction stir, a similar pattern of behaviour can be seen. Typically, the displacement/time graph can be divided into four phases.

Fig.3. Material displacement at the welding interface during welding
Fig.3. Material displacement at the welding interface during welding

In Phase I the parts are brought together and an axial force is applied. The interfacial friction begins but initially, no material flows. In Phase II, the weld zone material starts to melt and material displacement to the edges of the weld begins. Phase III is a steady state phase; the material is pushed out from the weld at a constant rate. Phase IV is the cooling phase when the interfacial friction is stopped but the force is still applied to consolidate the weld.

It is generally accepted that Phases I, II and IV are an essential part of the process but that there is no benefit, in terms of weld strength, in prolonging Phase III. Typically Phases I and II would take between 0.5 and 8 seconds to complete depending on the surface area being joined. Typically cooling times in Phase IV would be between 4 and 10 seconds.

Welding process parameters

In friction processes, welding can be carried out either until a pre-set weld time has elapsed or a pre-set material displacement has been achieved.

When welding by time, the weld time is the length of time the plastic parts are rubbed together to create the heat. As discussed previously, the weld time should ideally be terminated when the steady state phase of the weld cycle is achieved. This can be determined by using a displacement transducer. Higher melting point materials would typically require a longer weld time.

An alternative to welding by time is to weld by displacement. Interfacial friction is applied to components being welded until a fixed material displacement is achieved. This would typically be 1 to 2mm, but would depend on the flatness of the components being welded. Undulations in the welding interface would need to be taken into consideration when setting the weld displacement.

Applying a force to the component during welding creates a pressure at the joint interface. For friction welding of plastics, the typical welding and cooling pressure is between 0.5 and 2MPa. Increasing the weld pressure beyond these values can reduce the strength of the weld by forcing out most of the molten thermoplastic materials, resulting in a 'cold weld' being formed.

The cooling time is the length of time for which parts remain under pressure after the relative friction motion has ceased. Other welding process parameters are unique to the individual processes and include amplitude and frequency in the vibration process and rotational speed in the spin welding process.

Component design

Component design can be divided into the joint design and the design of the component itself. Joint and component design are critical to the success of friction welding processes, particularly in linear and orbital friction welding where flexing in the walls of the components can result in a reduction of the relative interfacial motion needed to produce friction heating. To eliminate this problem, it is important to include features such as stiffening ribs and U-flanges to the component wall around the weld area. The U-flange is particularly important since it is designed to lock the component wall to the tooling, thus preventing wall flexing. Wall flexing is especially a problem in vibration welding where the vibrations occur transverse to the wall of the component. Figure 4 shows a U-flange joint used in vibration welding, which can also be employed with other friction welding processes.

Fig.4. U-flange joint used in vibration or other friction welding processes
Fig.4. U-flange joint used in vibration or other friction welding processes


Friction welding processes are widely used techniques for the assembly of plastic components. Correct selection of welding parameters and component design are essential to successful welding using these processes.

More detailed information is available in the Friction welding of thermoplastics - a guide to best practice

See further information about plastics welding and testing or please contact us.

This Job Knowledge article was originally published in Connect, July - August 2002. It has been updated so the web page no longer reflects exactly the printed version.

For more information please email: