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

Subscribe >

Friction Stir Welding of Steel

Previous work at TWI has shown that FSW can produce high quality, low distortion welds in a wide range of steels, but the process has seen little industrial uptake in steel due to the short life and inconsistent performance of the tools then available. TWI Industrial Member, Element Six, has now entered the steel FSW tool market and provided a number of prototype tools for testing and evaluation at TWI. These were designed for welding 6mm thick steel.

Objectives

The objectives of the trials undertaken were to determine if:

  • The tools were capable of producing defect free welds in a carbon steel;
  • The tools behaved consistently;
  • The tools could make 40m of good weld without failing or wearing out.

Approach

The tools, two each from multiple production batches, were tested by making 6mm penetration, bead on plate welds in 8mm thick S355 steel. Bead on plate welds were made rather than butt welds to ensure greater consistency in testing, eliminating any variables that might arise from slight variations in plate edge alignment in a butt weld configuration. The welds were performed in position control and the surface of the steel was lightly milled before welding to remove the mill scale, again to eliminate any variations in condition between the plates used. Welding was undertaken at a speed of 300mm/min using TWI’s PowerStir™ FSW machine.

Three welding regimens were initially explored in the test programme:

  • Multiple, short (2m) long welds in which it was expected that service performance would be dominated by the effects of the initial plunge into cold steel;
  • Multiple, medium (5m) long welds, which were considered to represent a duty cycle likely to be seen in general engineering applications;
  • Multiple, long (20m) welds in which it was considered that tool performance would be dominated by thermal effects.

 

The tool lifetime and batch consistency element of the testing was undertaken by accumulating 40m of weld per tool using the 5m weld duty cycle, i.e. eight plunges and eight 5m long welds. A sample was taken from the end of each individual 5m weld to provide a macro-section, from which a qualitative assessment of weld quality could be made by examining for the presence of any flaws or change in the form of the welded region.

After the tools had each accumulated 40m of weld, a number of tools were then further tested to accumulate a weld length of 60m per tool, and these tools were used to make butt welds in S355 and DH36 steel. Samples were extracted from this butt weld for metallographic examination and mechanical testing (tensile strength and Charpy impact toughness).

A weld in progress at a speed of 300mm/min.
A weld in progress at a speed of 300mm/min.
Comparison of the microstructures of the parent steel, DH36 (left) and the weld zone, (right).
Comparison of the microstructures of the parent steel, DH36 (left) and the weld zone, (right).

Results

Tool life

The Element Six tools were capable of operating under all three different service regimens examined, and none exhibited an unpredictable, catastrophic failure. None of the tools wore sufficiently to generate defective welds over the 40m distance tested.

Metallography

Metallographic examination of the welds showed that, at the targeted 40m weld length, all the tools were capable of producing a fine grained weld with a mixed microstructure that was free from visible defects.

Weld strength

The cross weld tensile test failed in a ductile manner in the parent metal some distance from the weld and HAZ. This is consistent with previous experience of friction stir welds made in steel. The mean ultimate tensile strength of the tested samples was 562 MPa. No parent metal was tested but EN 10025-3 for S355 steel in the fine grained, rolled condition specifies a minimum UTS requirement of 520 MPa for steel less than 40mm in thickness. As the welded samples failed in the parent metal, it can therefore be stated that the weld strength was at least equal to 562 MPa and exceeded the minimum parent metal requirement.

Weld toughness

The Charpy impact testing was performed at -20oC and in accordance with BS EN ISO 148-1:2016. The samples were removed from mid weld thickness and notched in the welding direction through the thickness on the weld metal centreline. Impact strengths of 49, 57 and 61 Joules KV were recorded for the three test pieces, with the mean being 56J. This is considerably in excess of the minimum specified parent metal value for S355 at -20oC, which is 27 J KV.

Conclusions

The trials demonstrated that the tools made defect free welds under three different regimens:

  • Multiple 2m long welds, suitable for repair and applications such as pipeline girth welding;
  • Multiple 5m long welds, suitable for general engineering fabrication;
  • Multiple 20m long welds, suitable for applications such as shipbuilding.

 

The tools:

  • Have a minimum life of 40m of weld in 6mm thick S355 steel;
  • Exhibited consistent performance across multiple production batches;
  • Generated strong tough welds whose properties matched or exceeded that of the parent S355 steel.
An Element Six FSW tool before (left) and after (right) accumulating a total of 60m of weld.
An Element Six FSW tool before (left) and after (right) accumulating a total of 60m of weld.
An Element Six tool for FSW of 6mm thick steel prior to use (left) and after completing 40m of welding (right).
An Element Six tool for FSW of 6mm thick steel prior to use (left) and after completing 40m of welding (right).
Media /

Related Insights

View all insights
}