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

Subscribe >
Skip to content

What is a 'fish-eye' and how does it develop?


A 'fish-eye' is seen when a ferritic steel weld metal which contains hydrogen is plastically strained at a slow enough strain rate, at ambient temperature. The fish-eye comprises two regions - at the centre, the 'pupil' of the eye is a pore, inclusion, or other small defect. This is the only original imperfection associated with the fish-eye. The outer region, the 'iris', has fractured due to local hydrogen embrittlement in a quasi-cleavage mode, in a pattern radiating away from the 'pupil'.

There are several suggestions as to why fish-eyes form. One is that after welding, hydrogen collects in pores and in or near inclusions. When the weld is plastically strained, this hydrogen is released into the steel and causes embrittlement within a given radius of the defect. A more likely explanation is that when the bulk tensile strain is greater than the elastic limit of the hydrogen-embrittled weld metal, cracking begins at the largest defects present. As the amount of strain increases, the test piece begins to neck, which concentrates the strain and increases the strain rate in the necked region. The steel is no longer embrittled by hydrogen, as this only occurs at slow strain rates. Thus, the fish-eye crack is stopped at some distance from its origin.

It is important to test material with a hydrogen concentration representative of that which will be experienced in service, and this should override a desire to avoid fish-eyes. However, when the primary source of hydrogen is welding, and service loading will be some time after fabrication, it is reasonable to test after a hydrogen release treatment, or at least by leaving test specimens in a warm place for as long as possible between welding and testing. Fish-eyes are artefacts of testing a steel which contains hydrogen, and not defects in themselves.


1. Bailey, N.:"Fisheyes, hydrogen embrittlement and removal", TWI Research Bulletin, 15 (12), December 1974
2. Lundin, C and Patriarca, C:"Assessment of the significance of weld discontinuities: Effects of microstructure and discontinuities upon fracture morphology", Welding Research Council Bulletin 311, January 1986

See further information about Materials and Corrosion Management  or please contact us.

For more information please email: