Rotary friction welding (RFW) is a type of friction welding, which uses friction to heat two surfaces and create a non-separable weld. For rotary friction welding, this typically involves rotating one element relative to both the other element, and to the forge, while pressing them together with an axial force. This leads to the interface heating and then creating a permanent connection. Rotary friction welding can weld identical, dissimilar, composite, and non-metallic materials. Like other friction welding methods, it is a type of solid-state welding.
History
Rotary friction is the oldest of all friction welding methods, with a method of rotary friction welding first being patented in 1891. In 1956, the Russian machinist A. J. Chdikov (А. И. Чудиков), after having performed rotary friction welding with a lathe in the Elbrussky mine, would propose its commercial use to the Ministry of Metallurgy. While the Ministry of Metallurgy did not see value in this, it would attract attention from the national Scientific Research Institute of Electrical Welding Equipment, and was gradually disseminated following its publication in newspapers of the Soviet Union.
In 1960, the process would spread to the United States, with American companies such as Caterpillar Tractor Company (CAT), Rockwell International, and American Machine and Foundry developing machines for this process. This led to the development of an inertial friction welding process in 1962, through joint efforts from CAT and Manufacturing Technologies Incorporated (MTI). The 1960s also marked the first research of friction in welding in England by The Welding Institute (TWI). In Europe, KUKA AG and Thompson Friction Welding, would develop a direct-drive process and build a double spindle friction welder. The efficiency of friction welding, both linear and rotary, has been improved by the development of low force friction welding by the Edison Welding Institute and MTI working in collaboration.
Applications Rotary friction welding is widely implemented across the manufacturing sector and has been used for numerous applications, including:
Parts in gas turbines. Automotive parts Monel-to-steel marine fittings Cutting tools Tublar joints
Connections geometry Rotary Friction Welding can join a wide range of part geometries such as tube to tube, tube to disk, bar to plate. In addition, a rotating ring is used to connect long components. The geometry of the component surface does not have to be flat but can also be conical.
Types of materials to be welded Rotary friction welding enables to weld various materials.
Metallic materials of the same name or dissimilar either composite,superalloys and non-metallic e.g. thermoplastic polymers can be welded and even the welding of wood has been investigated. Weldability tables of metallic alloy can be found on the Internet and in books.
Sometimes, an interlayer is used to connect non-compatible materials.
Division due to drive motor In direct-drive friction welding (also called continuous drive friction welding), the drive motor and chuck are connected. The drive motor is continually driving the chuck during the heating stages. Usually, a clutch is used to disconnect the drive motor from the chuck, and a brake is then used to stop the chuck.
In inertia friction welding the drive motor is disengaged, and the workpieces are forced together by a friction welding force. The kinetic energy stored in the rotating flywheel is dissipated as heat at the weld interface as the flywheel speed decreases. Before welding, one of the workpieces is attached to the rotary chuck along with a flywheel of a given weight. The piece is then spun up to a high rate of rotation to store the required energy in the flywheel. Once spinning at the proper speed, the motor is removed and the pieces forced together under pressure. The force is kept on the pieces after the spinning stops to allow the weld to "set".Rotary Friction Welding, video and schematic diagram
Stages of process
Step 1 and 2, friction stage: one of the components is set in rotation, and then pressed to the other stationary one in axial of rotation, Step 3, braking stage: the rotating component is stopped in braking time, Step 4, upsetting stage: the welded elements are still forging by forge pressure (pressed down), Step 5: in standard RFW welding (standard parameters), a flash will be created. Outside flash can be cut off on the welder. However, referring to the stages chart:
modifications of the process exist, may depend on the version of the process: direct-drive, inertia friction welding, hybrid welding, there are many versions of welding machines, many materials can are welded with not the same properties, with various geometries, the real life process does not have to match to the ideal settings on the welding machine.
RFW Friction work on cylindrical rods workpieces
Friction work create weld and can believe that is calculated for cylindrical workpieces from math:
Work:
(1)
Moment of force M general formula:
(2)
The force F will be the frictional force T (F=T) so substituting for the formula (2):
(3)
The friction force T will be the pressure F times by the friction coefficient μ:
(4)
So moment of force M:
(5)
The alpha angle that each point will move with the axis of rotating cylindrical workpieces will be:
(6)
So friction work:
(7) [verification needed]
For variable value μ over friction time:
(8)
This requires verification but from the equation it appears that turnover and force (or pressure on surface ) is linear to friction work (W) so for example if the pressure increases 2 times then the friction work also increase 2 times, if the turnover increase 2 times then the friction work also increase 2 times and referring to conservation of energy this can heat 2 times the material to the same temperature or the temperature may increase 2 times. Pressure has the same effect over the entire surface but rotation has more impact away from the axis of rotation because it is a rotary motion. Referring to thermal conductivity the friction time affects to the flash size when shorter time was used then friction work is more concentrated in a smaller area.
or variable values μ, n, F over friction time:
(9)
t [s]- time of friction (when piece rotary), μ - coefficient of friction, F [N]- pressure force, r [m]- radius of workpiece, n [1/s] - turnover per second, W [J] - friction work.
Therefore, this way of calculating is not reliable in reality, which is more complicated. An example article considering the variable depends on the temperature coefficient of friction steel - aluminum Al60611 - Alumina is described by authors from Malaysia in for example this paper "Evaluation of Properties and FEM Model of the Friction Welded Mild Steel-Al6061-Alumina" and based on this position someone created no step by step but whatever an instructional simulation video in abaqus software and in this paper is possible to find the selection of the mesh type in the simulation described by the authors and there are some instructions such as use the Johnson-Cook material model choice, and not only, there is dissipation coefficient value, friction welding condition, the article included too the physical formulas related to rotary friction welding described by the authors such as: heat transfer equation and convection in rods, equations related to deformation processes. Article included information on the parameters of authors research, but it is not a step by step and simple instruction such as also the video and good add that it is not the only one position in literature. The conclusion include information that: "Even though the FE model proposed in this study cannot replace a more accurate analysis, it does provide guidance in weld parameter development and enhances understanding of the friction welding process, thus reducing costly and time consuming experimental approaches."
The coefficient of friction changes with temperature, and there are a number of other factors, including internal friction (measured in viscosity, e.g., dynamic viscosity according to Carreau's fluid law), forge, variable properties of the material during welding, and plastic deformation.
Carreau's fluid law:
Generalized Newtonian fluid where viscosity, , depends upon the shear rate, , by the following equation:
(10)
Where:
, , and are material coefficients. = viscosity at zero shear rate (Pa.s) = viscosity at infinite shear rate (Pa.s) = relaxation time (s) = power index Modelling of the frictional heat generated within the RFW process can be realized as a function of conducted frictional work and its dissipation coefficient. Incremental frictional work of a node 𝑖 on the contacting surface can be described as a function of its axial distance from the rotation centre, current frictional shear stress, rotational speed and incremental time. The dissipation coefficient 𝛽FR is often set to 0.9 meaning that 90% of frictional work is dissipated into heat.
(11) 𝑑𝑞FR(𝑖) = 𝛽FR ∙ 𝑑𝑊FR(𝑖) = 𝛽FR ∙ 𝜏𝑅(𝑖) ∙ 𝜔 ∙ 𝑟𝑖 ∙ 𝑑𝑡 on contacting surface of node 𝑖
𝛽FR - dissipation coefficient, 𝑊FR - frictional work, 𝑟𝑖 - distance from the rotation centre, dt - time increment, 𝜏𝑅(𝑖) - current frictional shear stress, 𝜔 - rotational speed.
Friction work can also calculate from power of used for welding and friction time (will not be greater than the friction time multiply to the power of the welder - engine of the welder) referring to rules conservation of energy. This calculation looks the simplest.
(12) E = Pxt or for not constant power
E - energy, P - power, t - power runtime.
However, in this case, energy can be also stored in the flywheel if is used depending on the welder construction.
General flywheel energy formula:
(13)
where:
is the stored kinetic energy, ω is the angular velocity, and is the moment of inertia of the flywheel about its axis of symmetry. Sample calculations not by computer simulation also exist in the literature, for example, in a 1974 script related to power input and temperature distribution.
K. K. Wang and Wen Lin from Cornell University in "Flywheel friction welding research" manually calculates welding process, and even at this time, the weld structure was analysed.
Overall, calculations can be complex.
Weld Zone Description
Weld photo gallery +
Heat and mechanical affected zones Friction work is converted into increased temperatures at the welding zone, and as a result, the weld structure is changed. In typical rotary friction welding processes, rise of temperature at the beginning of processes should be further from the axis of rotation, because points further from the axis have greater linear velocity. During the weld, the temperature disperses according to the thermal conductivity of the welded parts.
Zones:
WCZ—weld center zone, HAZ—heat affected zone, TMAZ—Thermo-Mechanically Affected Zone, BM—base material, parent material, Flash
Both the WCZ and the TMAZ are thermo-mechanically affected. However, due to differences in the microstructures, they are classified independently. One such difference is that only the WCZ undergoes dynamic recrystallization. The Plastically Affected Zone (PAZ) is the region between the two TMAZ-HAZ boundaries, also known as the TMAZ thickness. The HAZ can also be subdivided based on temperature differences within it.
Next, the heat spreads, and the material is pushed out, creating a flash which can be cut off on the welding machine.
Weld measuring system To provide knowledge about the process, monitoring systems are often used and this are carried out in several ways which affects the accuracy and the list of measured parameters.
The list of measured and calculated parameters can looks like this:
axial force and pressure, angular - rotation speed, spindle centre, velocity, vibration, length (burn off rate), temperature.