ZHUOJIN-LOGOZhuojin Pipe Fitting Co., Ltd
Zhuojin Pipe Fitting Co., Ltd
STEEL PIPE · FITTING · WELDING  |  NOTES FROM A 30-YEAR INDUSTRY VETERAN

Finite Element Simulation of Strip Submerged Arc Hardfacing Temperature Fields on Thick Plates

Why Simulate a Welding Temperature Field?

The paper by Wang Zhifeng and colleagues from Harbin Welding Research Institute and China First Heavy Industries is a technical study on the finite element simulation of the temperature field during strip submerged arc hardfacing on thick plates. This is a topic that might seem academic, but it has direct practical implications for anyone who runs strip cladding operations, which is a common method for surfacing pressure vessels, reactor vessels, and other thick-walled components with corrosion-resistant alloys.

I have run strip cladding lines for years, and I can tell you that the biggest challenge is controlling the dilution rate and the shape of the weld bead. If the heat input is too high, you get excessive dilution, which means the base metal mixes into the cladding and degrades its corrosion resistance. If the heat input is too low, you get lack of fusion and porosity. The temperature field determines the dilution, the bead shape, and the residual stresses, so being able to predict it is valuable.

The paper uses the Goldak heat source model, which is a double ellipsoidal heat source that represents the moving arc. This is a well-established model, but the authors adapt it for strip electrodes, which have a rectangular cross-section rather than a circular one. This adaptation is the key contribution of the paper.

The Heat Source Model and Simulation Approach

The authors designed a finite element heat source model based on the strip submerged arc hardfacing principle and the Goldak model theory. They implemented it on the MSC.MARC platform using a FORTRAN subroutine to load the moving heat source. The simulation was validated by comparing the calculated weld bead shape and thermal cycle curves with experimental measurements.

Parameter Value / Description Why It Matters
Heat source model Double ellipsoidal (Goldak), adapted for strip electrode Represents the rectangular heat distribution of a strip
Simulation platform MSC.MARC with FORTRAN subroutine Allows moving heat source loading
Validation method Weld bead shape and thermal cycle curves Confirms the model accuracy
Plate thickness Thick plate, likely 100 mm or more Represents pressure vessel and reactor applications
Strip electrode width Not specified, but typical strip is 30-60 mm wide Determines the heat distribution width

The Goldak model is a good choice because it allows the heat flux distribution to be adjusted in the x, y, and z directions. For a strip electrode, the heat flux is not circular but rectangular, so the model needs to be modified to account for the width of the strip. The authors did this by adjusting the parameters of the double ellipsoid to match the strip geometry.

The use of MSC.MARC with a FORTRAN subroutine is a standard approach for moving heat source simulations. The subroutine defines the heat flux as a function of position and time, and it is called at each time step to apply the heat load. This is computationally intensive but accurate.

Simulation Results and Practical Implications

The simulation results showed good agreement with the experimental measurements, both in terms of the weld bead shape and the thermal cycle curves. This is important because it means the model can be used to predict the temperature field for different welding parameters without having to run expensive and time-consuming experiments.

From a practical standpoint, the temperature field tells you several things:

  1. The peak temperature at the interface between the cladding and the base metal determines the dilution rate. If the peak temperature is too high, the base metal melts excessively, and dilution increases. The simulation can help you optimize the welding current and travel speed to achieve the desired dilution.
  2. The cooling rate in the heat-affected zone determines the microstructure and hardness of the base metal. For a thick plate, the cooling rate is relatively slow, which can lead to grain growth and softening. The simulation can help you predict the HAZ width and the resulting properties.
  3. The residual stresses after welding are determined by the temperature field and the subsequent cooling. The simulation can be extended to a mechanical analysis to predict the residual stress distribution, which is critical for pressure vessel applications where stress corrosion cracking is a concern.

I want to add a practical observation from my own experience. The simulation is a valuable tool, but it is only as good as the input parameters. The heat input, the thermal conductivity, the specific heat, and the convective heat transfer coefficients all need to be accurate. In my experience, the convective heat transfer coefficient is the most uncertain parameter, and it can vary significantly depending on the shielding gas flow, the ambient temperature, and the surface condition. I always recommend validating the simulation with a few experimental measurements before using it for optimization.

Limitations and My Judgment

The paper is a solid contribution to the field of welding simulation. The adaptation of the Goldak model for strip electrodes is a useful step forward, and the validation against experimental data gives confidence in the results.

However, I want to point out a few limitations. First, the simulation is a thermal analysis only; it does not include the mechanical effects of welding, such as residual stress and distortion. For a complete understanding of the cladding process, you would need to couple the thermal and mechanical analyses. Second, the model assumes a constant heat input, but in reality, the arc efficiency and the heat distribution can vary along the weld. Third, the simulation does not account for the effect of the molten pool dynamics, which can affect the bead shape.

My practical judgment is that this simulation is most useful for parametric studies, such as optimizing the welding current, voltage, and travel speed to achieve a target dilution rate or bead shape. It is less useful for predicting the absolute temperature at a specific point, because the uncertainties in the input parameters can lead to significant errors. I would use it as a guide, but I would always confirm the results with a few experimental welds.

The paper also reminds me of the importance of understanding the physics behind the process. A simulation is not a substitute for experience; it is a tool that enhances experience. The best welding engineers are those who can look at a simulation result and immediately know whether it makes sense based on their own observations of the actual weld.