Map each tank-shunt packet’s rolling direction onto its intended local flux path. Carry the material axes through cutting and assembly so the mounted steel delivers the assessed magnetic orientation. [1]
Trace the collected flux through the packet
Start with the leakage-field source and the flux entering the shunt surface. Follow the path inside the packet toward its ends, corners and any connection to another packet. The direction can change within that path. A layout aligned well in the central region may send part of the flux across the rolling direction at a turn.
Keep rolling direction, in-plane transverse direction and the direction through the laminated stack distinct. Coating gaps and stacking also affect the through-stack path. Substituting a high rolling-direction permeability in all three directions can create an unrealistically attractive collector in the model.
| Shunt region | Direction record | Check required |
|---|---|---|
| Central collecting span | Rolling direction relative to dominant internal flux | Representative magnetization curve and operating range |
| Corner or turn | Changing in-plane flux direction | Off-axis response and local field concentration |
| Packet interface | Material axes on both sides | Gap and transfer path across the interface |
| Outer packet | Axes relative to edge-fringing flux | Local loading beyond the central average |
Register the axes in the model
Assign a local material frame to each packet and transform it consistently into the model coordinates. In a linear illustrative formulation, a permeability tensor in material coordinates is rotated into the model frame with the corresponding rotation matrix. For nonlinear GOES, that bookkeeping does not replace a validated directional constitutive description.
Manufacturer descriptions establish that GOES magnetic performance is directional. The tank-shield field paper also shows that collected flux is nonuniform across the assembly. Combining those observations justifies a local orientation check; neither source supplies a universal optimum for every shunt layout. [1]
Compare local corner response alongside total loss
Hold packet geometry, incident excitation and material identity constant when comparing orientation arrangements. Inspect local flux density and field requirement at turns and interfaces, then examine the tank and shunt loss distribution with an appropriate loss model. A small total-loss difference can conceal a new concentrated region.
Nippon’s 0.23 mm 23ZH90 catalogue gives longitudinal loss ≤0.90 W/kg at 1.7 T, 50 Hz [2]. Assume a 100 mm rolling-direction segment and a 40 mm transverse segment. A fictional 2 kg packet would have a 1.80 W longitudinal strip budget only if its test condition applied throughout. The transverse route occupies 28.57% of length, but multiplying 1.80 W by that fraction cannot estimate its contribution: direction changes H and local flux, not just steel allocation. Reject a model applying the longitudinal limit to the whole packet. Obtain directional data or change the corner/overlap so its unsupported transverse duty is reduced.
Cutting orientation, bend/overlap geometry and packet labels establish rolling direction in the installed shunt. Inspect those features against the field-path drawing before closing the tank.
Confirm the as-built rolling-direction map during fabrication and assembly. Preserve it in procurement records with the direction-dependent material evidence used for the local flux assessment.
References
- Z. Jurković, B. Jurišić and T. Župan, Multi-Step Approach for Fast Calculation of Magnetic Field in Transformer Tank Shields, Energies 17, 1378, 2024. DOI: 10.3390/en17061378.
- Nippon Steel Corporation, Grain-Oriented Electrical Steel Sheets, catalogue D004je, April 2020, specification and typical-property tables.