Manual vs. robotic weld profiling of monopile and tower girth welds
Profiling of inside and outside circumferential/girth weld seams in monopile foundations and tower sections may be performed either manually, using belt or disc grinders, or using the Maxars TM401 weld profiling robot system.
Both aim to meet DNV-RP-C203’s target geometry: a large-radius transition from weld toe to base plate with complete undercut removal within the depth limits specified by DNV-RP-C203.
Where manual grinding relies on operator skill to meet these geometry improvement recommendations, CNC-robotic weld profiling achieves it through programmed and laser-guided machining operation with a dust-free and repeatable-quality (uniform-profile) outcome.
DNV-RP-C203 requirements for weld profiling of monopiles and wind turbine towers
DNV recommended practice DNV-RP-C203, Fatigue design of offshore steel structures, governs the profiling of inside and outside circumferential (girth) butt-weld seams in monopile foundation and tower section manufacturing.
DNV-RP-C203 defines the target outcome rather than the tooling: the weld cap is to be removed and blended into a large-radius, continuous transition from weld seam to base plate, and notched/weld toe undercuts introduced by submerged arc welding (SAW) processes are to be removed to a depth below the visible undercut of 0.5 mm where manual grinding is used, or 0.2 mm where machining is used.
Total material removal must not exceed 2 mm or 7% of plate thickness, whichever is smaller.
Manual weld grinding: slow, labour-intensive and operator-dependent
The established method for applying weld profiling is manual grinding using a disc or belt grinder. This is a slow, labour-intensive process: typical manual grinding rates are in the order of 2-3 metres of weld seam per hour.
The more significant limitation is consistency. The International Institute of Welding (IIW) has noted that the fatigue strength improvement achieved by manual grinding varies widely from weld to weld, and that this variation is driven largely by the skill of the individual operator.
Independent findings by Smith and Hirt add that grinding defects introduced during manual profiling can be as serious as the defects being removed, and that manual grinding frequently falls short of fully addressing weld toe undercuts as recommended by DNV-RP-C203.
In other words, manual grinding applies the same recommended practice and target profile geometry as CNC-robotic weld profiling: large toe radius, undercuts fully removed, and profiling depth held within the specified limits. The degree to which it meets target geometry depends heavily on the skill and focus of the person operating the grinder that day.
CNC-robotic weld profiling with the Maxars TM401: 30–60 m/h at ±0.2 mm accuracy
The Maxars MAX TM401 is a mobile 6-axis CNC robot featuring custom PLC software, precision hydraulics and motion control for automatic XYZ-positioning of the machining tool to ±0.2 mm.
The TM401 ensures consistency and delivers a repeatable-quality weld profiling outcome by replacing a skill-dependent manual process with laser-guided robotic weld profiling.
Acquisition, characterisation tracking and profiling of weld seams is performed automatically and according to input and programme and the TM401 performs single-pass flush (large-radius) weld profiling of inside and outside girth welds in wind turbine support structures at a rate of 30-60 meters/hour.
Adjustment for plate off-sets up to ±3 mm is automatic and profiling of thickness transition joints (both with and without taper) is according to weld seam selection and programme.
Standard uniform profiling result is a slightly concave profile with a 10 mm radius transition from weld toe to plate and a repeatable surface roughness below Ra 3.2 µm.
Why consistent weld profiling matters for fatigue design under DNV-RP-C203
The International Institute of Welding (IIW) has pointed out that there are large variations in the actual fatigue strength improvements achieved by manual weld grinding, and that one reason for this variation is that the effectiveness of manual weld profiling depends heavily on the skill of the operator.
The large variations in the actual fatigue strength improvements achieved by manual weld profiling introduce uncertainty because these large variations indicate that the benefits of manual weld profiling are difficult to predict accurately.
This uncertainty creates risk because of the challenge it presents in preparing or mitigating potential negative impacts. If designers of monopile foundations and wind turbine towers are uncertain about the outcome and benefits of manual weld profiling, they will be unable to accurately assess the risk associated and therefore cannot apply the theoretical benefits of weld profiling when doing fatigue design on these structures.
By delivering consistent and documented weld profiling and fatigue strength improvement on every weld, the TM401 is positioned to let DNV-RP-C203 fatigue benefit be applied with greater confidence. A distinction that becomes more consequential with the October 2024 edition of DNV-RP-C203, whose new Appendix F.16 curves for high-quality butt welds assume tighter fabrication tolerances (2–3 mm misalignment and defined weld cap flank angle limits) than manual practice has traditionally delivered as standard.
A closer comparison of quality, productivity, and working environment between manual and CNC-robotic weld profiling is covered in a separate FAQ.
Sources: DNV-RP-C203, Fatigue design of offshore steel structures (October 2024 edition), DNV. P. J. Haagensen & S. J. Maddox, IIW Recommendations on Post Weld Improvement of Steel and Aluminium Structures, revised 16 February 2004. Smith, I.J. & Hirt, M.A., findings on defects introduced by manual grinding. Maxars TM401 Operational Performance and Technical Data Sheet.