A core packet with several conductive contacts can contain alternative closed paths. If one bridge opens, one loop may disappear while another remains or carries a different current. The change is a network transition, not necessarily complete fault elimination.
Record the original contact state before interpreting the later measurement. Unclamping, moving debris or repairing an edge can alter more than one connection. A simplified before/after drawing should mark known changes and uncertain contacts separately.
Rebuild the return paths
Represent conductive sheet segments and contacts as a connected network. Trace closed paths after the identified connection changes. Include any verified external returns that belong to the investigation. Paths without evidence should remain hypotheses rather than silently added branches.
For each surviving path, consider linked flux and electrical impedance. Current redistribution depends on those quantities, not only on graph connectivity. Shared segments and magnetic coupling can make isolated-loop calculations inadequate.
| After-state observation | Candidate interpretation | Discriminating evidence |
|---|---|---|
| One local hotspot disappears | Its source path changed | Check remaining returns and registered source region |
| Another region warms more | Current or thermal boundary redistributed | Electrical response plus cooling-state comparison |
| Total loss falls but local heating remains | Some heating removed, another source persists | Local circuit and thermal mapping |
| No difference is resolved | Change may be below capability or path persists | Measurement sensitivity and contact-state verification |
A lower total-loss value should not be used to erase a remaining local question.
Keep the after-state comparable
Use the same excitation and measurement temperature where the comparison requires them. If the contact change also changes packet restraint or cooling, document those effects. A temperature difference can arise from a different thermal path even when electrical heating changes little.
Register local measurements to the same coordinates and preserve their timing. The hottest point can move after redistribution, so measuring only the former maximum may miss the new location.
Test the network explanation
Compare predicted changes in accessible electromagnetic and thermal observations with the actual after-state. A model that predicts current transfer to another branch should identify an observable consequence, rather than merely fit the reduced aggregate power.
Where contacts are nonlinear or intermittent, retain the relevant pressure and motion history. An open path observed in a static electrical check may reappear under the assembled condition that created the original heating.
The investigation is complete when the remaining network and its functional behavior are adequately resolved for the decision. That may establish elimination of the relevant loop, identify a surviving path, or require further observation. The before/after contact map keeps those outcomes distinct and prevents one vanished hotspot from becoming an unsupported whole-packet repair claim.
Recalculate the branch that survives a bridge change
Use 0.23 mm nominal sheets [1] and assume a loop area of 100 sheet thicknesses × 100 mm = 2,300 mm². A uniform normal field of 0.50 T peak at 50 Hz induces 0.2555 V rms. Assume a common 0.10 Ω segment followed by two parallel 0.10 Ω return branches, with negligible inductance. Before opening one branch, total current is 1.703 A and each return carries 0.852 A; after opening it, the surviving return carries 1.277 A. Its heating rises from 0.0725 W to 0.163 W, a factor of 2.25, even though total network power falls. Check topology and impedance uncertainty [2] before interpreting one vanished hotspot as complete repair. Faraday induction supplies the voltage relation [3]. The assumed field is normal to this loop; main rolling-direction flux is not automatically linked.
Manufacturing controls. Edge bridges, conductive debris, coating damage and clamp-dependent contacts set network connectivity. A repair record must identify all changed paths and compare the full registered region, including possible new maxima.
References
- Nippon Steel Corporation, Grain-Oriented Electrical Steel Sheets, catalogue D004je_03_202004, 2020, Specifications; Size and Form Tolerances.
- JCGM, JCGM 100:2008, Evaluation of measurement data — Guide to the expression of uncertainty in measurement, 2008.
- Zhao, X. et al., Magnetostrictive Properties of the Grain-Oriented Silicon Steel Sheet under DC-Biased and Multisinusoidal Magnetizations, Materials 12, 2156, 2019, sections 2.1 and 2.3; DOI: 10.3390/ma12132156.